AI Research News Feeds for September 30th, 2025

AI RESEARCH PAPERS & ACADEMIC SOURCES

• Responsible AI Technical Report
• Embedding Domain Knowledge for Large Language Models via Reinforcement Learning from Augmented Generation
• Beyond Sharp Minima: Robust LLM Unlearning via Feedback-Guided Multi-Point Optimization
• EmbeddingGemma: Powerful and Lightweight Text Representations
• i-LAVA: Insights on Low Latency Voice-2-Voice Architecture for Agents
• SiNGER: A Clearer Voice Distills Vision Transformers Further
• Can Less Precise Be More Reliable? A Systematic Evaluation of Quantization's Impact on CLIP Beyond Accuracy
• Towards Foundation Models for Zero-Shot Time Series Anomaly Detection: Leveraging Synthetic Data and Relative Context Discrepancy
• Evaluating the Effectiveness of Transformer Layers in Wav2Vec 2.0, XLS-R, and Whisper for Speaker Identification Tasks
• TReF-6: Inferring Task-Relevant Frames from a Single Demonstration for One-Shot Skill Generalization
• Multimodal Iterative RAG for Knowledge-Intensive Visual Question Answering
• Can General-Purpose Omnimodels Compete with Specialists? A Case Study in Medical Image Segmentation
• GradES: Significantly Faster Training in Transformers with Gradient-Based Early Stopping
• Do LLMs Adhere to Label Definitions? Examining Their Receptivity to External Label Definitions
• Co-Evolving Complexity: An Adversarial Framework for Automatic MARL Curricula
• Diffusion Generative Models Meet Compressed Sensing, with Applications to Imaging and Finance
• The Physical Basis of Prediction: World Model Formation in Neural Organoids via an LLM-Generated Curriculum
• BranchGRPO: Stable and Efficient GRPO with Structured Branching in Diffusion Models
• COMPACT: Common-token Optimized Model Pruning Across Channels and Tokens
• DEPFusion: Dual-Domain Enhancement and Priority-Guided Mamba Fusion for UAV Multispectral Object Detection
• A Systematic Survey on Large Language Models for Evolutionary Optimization: From Modeling to Solving
• TalkPlayData 2: An Agentic Synthetic Data Pipeline for Multimodal Conversational Music Recommendation
• FuseCodec: Semantic-Contextual Fusion and Supervision for Neural Codecs
• MindVL: Towards Efficient and Effective Training of Multimodal Large Language Models on Ascend NPUs
• Learning to Route: Per-Sample Adaptive Routing for Multimodal Multitask Prediction
• Evaluating undergraduate mathematics examinations in the era of generative AI: a curriculum-level case study
• TreeIRL: Safe Urban Driving with Tree Search and Inverse Reinforcement Learning
• WorldForge: Unlocking Emergent 3D/4D Generation in Video Diffusion Model via Training-Free Guidance
• Fast and Fluent Diffusion Language Models via Convolutional Decoding and Rejective Fine-tuning
• Accurate and Efficient Low-Rank Model Merging in Core Space
• StefaLand: An Efficient Geoscience Foundation Model That Improves Dynamic Land-Surface Predictions
• Joint Memory Frequency and Computing Frequency Scaling for Energy-efficient DNN Inference
• Reinforced Generation of Combinatorial Structures: Applications to Complexity Theory
• Self-Evolving LLMs via Continual Instruction Tuning
• APRIL: Active Partial Rollouts in Reinforcement Learning to Tame Long-tail Generation
• SPiDR: A Simple Approach for Zero-Shot Safety in Sim-to-Real Transfer
• Diversity Boosts AI-Generated Text Detection
• Model Collapse Is Not a Bug but a Feature in Machine Unlearning for LLMs
• PRIME: Large Language Model Personalization with Cognitive Dual-Memory and Personalized Thought Process
• CoSteer: Collaborative Decoding-Time Personalization via Local Delta Steering
• Entropy-Memorization Law: Evaluating Memorization Difficulty of Data in LLMs
• Mitigating Watermark Forgery in Generative Models via Randomized Key Selection
• Function Induction and Task Generalization: An Interpretability Study with Off-by-One Addition
• BenchRL-QAS: Benchmarking reinforcement learning algorithms for quantum architecture search
• Vidar: Embodied Video Diffusion Model for Generalist Manipulation
• Making Language Model a Hierarchical Classifier
• Learning to summarize user information for personalized reinforcement learning from human feedback
• GRID: Scalable Task-Agnostic Prompt-Based Continual Learning for Language Models
• Omni-Thinker: Scaling Multi-Task RL in LLMs with Hybrid Reward and Task Scheduling
• The Ever-Evolving Science Exam
• When Engineering Outruns Intelligence: Rethinking Instruction-Guided Navigation
• Can Language Models Discover Scaling Laws?
• Communicating Plans, Not Percepts: Scalable Multi-Agent Coordination with Embodied World Models
• The Geometry of Cortical Computation: Manifold Disentanglement and Predictive Dynamics in VCNet
• Sculptor: Empowering LLMs with Cognitive Agency via Active Context Management
• AttriLens-Mol: Attribute Guided Reinforcement Learning for Molecular Property Prediction with Large Language Models
• PakBBQ: A Culturally Adapted Bias Benchmark for QA
• BLADE: Block-Sparse Attention Meets Step Distillation for Efficient Video Generation
• ECHO: Frequency-aware Hierarchical Encoding for Variable-length Signals
• CORE-RAG: Lossless Compression for Retrieval-Augmented LLMs via Reinforcement Learning
• Automatic Question & Answer Generation Using Generative Large Language Model (LLM)
• End-to-End On-Device Quantization-Aware Training for LLMs at Inference Cost
• Improving LLM Reasoning through Interpretable Role-Playing Steering
• InverseScope: Scalable Activation Inversion for Interpreting Large Language Models
• TGRPO :Fine-tuning Vision-Language-Action Model via Trajectory-wise Group Relative Policy Optimization
• Towards Robust Real-World Multivariate Time Series Forecasting: A Unified Framework for Dependency, Asynchrony, and Missingness
• When Is Diversity Rewarded in Cooperative Multi-Agent Learning?
• Revisiting Visual Understanding in Multimodal Reasoning through a Lens of Image Perturbation
• Beyond Jailbreaking: Auditing Contextual Privacy in LLM Agents
• Discrete Audio Tokens: More Than a Survey!
• Enhancing Delta Compression in LLMs via SVD-based Quantization Error Minimization
• Meta Pruning via Graph Metanetworks : A Universal Meta Learning Framework for Network Pruning
• StorySage: Conversational Autobiography Writing Powered by a Multi-Agent Framework
• Decoupled Classifier-Free Guidance for Counterfactual Diffusion Models
• Long-Context Generalization with Sparse Attention
• Do We Need Large VLMs for Spotting Soccer Actions?
• Adaptive Sample Scheduling for Direct Preference Optimization
• From Drawings to Decisions: A Hybrid Vision-Language Framework for Parsing 2D Engineering Drawings into Structured Manufacturing Knowledge
• GRAF: Multi-turn Jailbreaking via Global Refinement and Active Fabrication
• Improving Black-Box Generative Attacks via Generator Semantic Consistency
• OmniGen2: Exploration to Advanced Multimodal Generation
• Reasoning Isn't Enough: Examining Truth-Bias and Sycophancy in LLMs
• R1-Ranker: Teaching LLM Rankers to Reason
• Enhancing Live Broadcast Engagement: A Multi-modal Approach to Short Video Recommendations Using MMGCN and User Preferences
• Semantic-guided Diverse Decoding for Large Language Model
• Data Uniformity Improves Training Efficiency and More, with a Convergence Framework Beyond the NTK Regime
• Theoretical Modeling of LLM Self-Improvement Training Dynamics Through Solver-Verifier Gap
• Learning to Segment for Vehicle Routing Problems
• Empirical Analysis Of Heuristic and Approximation Algorithms for the The Mutual-Visibility Problem
• Latent Chain-of-Thought? Decoding the Depth-Recurrent Transformer
• Communication-Efficient Desire Alignment for Embodied Agent-Human Adaptation
• CryoCCD: Conditional Cycle-consistent Diffusion with Biophysical Modeling for Cryo-EM Synthesis
• ReCalKV: Low-Rank KV Cache Compression via Head Reordering and Offline Calibration
• Improving Reliability and Explainability of Medical Question Answering through Atomic Fact Checking in Retrieval-Augmented LLMs
• ProxyThinker: Test-Time Guidance through Small Visual Reasoners
• TRAPDOC: Deceiving LLM Users by Injecting Imperceptible Phantom Tokens into Documents
• WorldGym: World Model as An Environment for Policy Evaluation
• GRAM: Spatial general-purpose audio representation models for real-world applications
• Angles Don't Lie: Unlocking Training-Efficient RL Through the Model's Own Signals
• Interaction Field Matching: Overcoming Limitations of Electrostatic Models
• VisCoder: Fine-Tuning LLMs for Executable Python Visualization Code Generation
• Bridging the Performance Gap Between Target-Free and Target-Based Reinforcement Learning
• Towards Better Generalization via Distributional Input Projection Network
• Reshaping Reasoning in LLMs: A Theoretical Analysis of RL Training Dynamics through Pattern Selection
• TreeRPO: Tree Relative Policy Optimization
• ReactDance: Hierarchical Representation for High-Fidelity and Coherent Long-Form Reactive Dance Generation
• OnPrem.LLM: A Privacy-Conscious Document Intelligence Toolkit
• SALM: A Multi-Agent Framework for Language Model-Driven Social Network Simulation
• How Hungry is AI? Benchmarking Energy, Water, and Carbon Footprint of LLM Inference
• Fine-tuning Diffusion Policies with Backpropagation Through Diffusion Timesteps
• Learning When to Think: Shaping Adaptive Reasoning in R1-Style Models via Multi-Stage RL
• Visual Planning: Let's Think Only with Images
• Stepwise Guided Policy Optimization: Coloring your Incorrect Reasoning in GRPO
• Continuous Optimization for Feature Selection with Permutation-Invariant Embedding and Policy-Guided Search
• Fine-grained Contrastive Learning for ECG-Report Alignment with Waveform Enhancement
• AdaBoN: Adaptive Best-of-N Alignment
• VFRTok: Variable Frame Rates Video Tokenizer with Duration-Proportional Information Assumption
• MobileIPL: Enhancing Mobile Agents Thinking Process via Iterative Preference Learning
• Is Active Persona Inference Necessary for Aligning Small Models to Personal Preferences?
• Mechanistic Fine-tuning for In-context Learning
• Language Models Optimized to Fool Detectors Still Have a Distinct Style (And How to Change It)
• PiFlow: Principle-aware Scientific Discovery with Multi-Agent Collaboration
• ReflAct: World-Grounded Decision Making in LLM Agents via Goal-State Reflection
• Scaling Diffusion Transformers Efficiently via $\mu$P
• Better Safe Than Sorry? Overreaction Problem of Vision Language Models in Visual Emergency Recognition
• Advancing Marine Research: UWSAM Framework and UIIS10K Dataset for Precise Underwater Instance Segmentation
• Leveraging Online Data to Enhance Medical Knowledge in a Small Persian Language Model
• Not All Models Suit Expert Offloading: On Local Routing Consistency of Mixture-of-Expert Models
• Scalable Graph Generative Modeling via Substructure Sequences
• AdaSTaR: Adaptive Data Sampling for Training Self-Taught Reasoners
• BitHydra: Towards Bit-flip Inference Cost Attack against Large Language Models
• Runtime Adaptive Pruning for LLM Inference
• InfoDet: A Dataset for Infographic Element Detection
• On the Design of KL-Regularized Policy Gradient Algorithms for LLM Reasoning
• Reward Model Overoptimisation in Iterated RLHF
• Improved Sample Complexity For Diffusion Model Training Without Empirical Risk Minimizer Access
• ThanoRA: Task Heterogeneity-Aware Multi-Task Low-Rank Adaptation
• LoTA-QAF: Lossless Ternary Adaptation for Quantization-Aware Fine-Tuning
• EnvSDD: Benchmarking Environmental Sound Deepfake Detection
• A Necessary Step toward Faithfulness: Measuring and Improving Consistency in Free-Text Explanations
• ePC: Overcoming Exponential Signal Decay in Deep Predictive Coding Networks
• Variational Deep Learning via Implicit Regularization
• GLEAM: Learning Generalizable Exploration Policy for Active Mapping in Complex 3D Indoor Scenes
• PDFBench: A Benchmark for De novo Protein Design from Function
• Any-to-Bokeh: Arbitrary-Subject Video Refocusing with Video Diffusion Model
• Cross-modal RAG: Sub-dimensional Text-to-Image Retrieval-Augmented Generation
• Half-order Fine-Tuning for Diffusion Model: A Recursive Likelihood Ratio Optimizer
• Advanced Architectures Integrated with Agentic AI for Next-Generation Wireless Networks
• A Statistical Learning Perspective on Semi-dual Adversarial Neural Optimal Transport Solvers
• LEAD: Large Foundation Model for EEG-Based Alzheimer's Disease Detection
• 3D Foundation Model for Generalizable Disease Detection in Head Computed Tomography
• Towards Large-Scale In-Context Reinforcement Learning by Meta-Training in Randomized Worlds
• UltraIF: Advancing Instruction Following from the Wild
• Confidence Improves Self-Consistency in LLMs
• OrderFusion: Encoding Orderbook for End-to-End Probabilistic Intraday Electricity Price Forecasting
• Comprehensive Review of Neural Differential Equations for Time Series Analysis
• Collaborative Deterministic-Probabilistic Forecasting for Diverse Spatiotemporal Systems
• PAFT: Prompt-Agnostic Fine-Tuning
• B-cos LM: Efficiently Transforming Pre-trained Language Models for Improved Explainability
• Mitigating Barren Plateaus in Quantum Neural Networks via an AI-Driven Submartingale-Based Framework
• MemeIntel: Explainable Detection of Propagandistic and Hateful Memes
• Mixing Any Cocktail with Limited Ingredients: On the Structure of Payoff Sets in Multi-Objective POMDPs and its Impact on Randomised Strategies
• SRA-CL: Semantic Retrieval Augmented Contrastive Learning for Sequential Recommendation
• Delta-Triplane Transformers as Occupancy World Models
• UniF$^2$ace: A Unified Fine-grained Face Understanding and Generation Model
• How to Protect Yourself from 5G Radiation? Investigating LLM Responses to Implicit Misinformation
• Implicit Bias-Like Patterns in Reasoning Models
• RISE: Robust Imitation through Stochastic Encoding
• What Makes a Reward Model a Good Teacher? An Optimization Perspective
• On The Sample Complexity Bounds In Bilevel Reinforcement Learning
• Reasoning to Learn from Latent Thoughts
• Machine Learning - Driven Materials Discovery: Unlocking Next-Generation Functional Materials - A review
• AdaRank: Adaptive Rank Pruning for Enhanced Model Merging
• SUV: Scalable Large Language Model Copyright Compliance with Regularized Selective Unlearning
• XL-Suite: Cross-Lingual Synthetic Training and Evaluation Data for Open-Ended Generation
• Beyond Synthetic Replays: Turning Diffusion Features into Few-Shot Class-Incremental Learning Knowledge
• SentenceKV: Efficient LLM Inference via Sentence-Level Semantic KV Caching
• SCRAMBLe : Enhancing Multimodal LLM Compositionality with Synthetic Preference Data
• From Specificity to Generality: Revisiting Generalizable Artifacts in Detecting Face Deepfakes
• Optuna vs Code Llama: Are LLMs a New Paradigm for Hyperparameter Tuning?
• Min-Max Optimisation for Nonconvex-Nonconcave Functions Using a Random Zeroth-Order Extragradient Algorithm
• Efficient Reasoning Models: A Survey
• IPBench: Benchmarking the Knowledge of Large Language Models in Intellectual Property
• Dynamic Early Exit in Reasoning Models
• Evolution Meets Diffusion: Efficient Neural Architecture Generation
• Cooking Up Creativity: Enhancing LLM Creativity through Structured Recombination
• $\textit{New News}$: System-2 Fine-tuning for Robust Integration of New Knowledge
• Beyond Losses Reweighting: Empowering Multi-Task Learning via the Generalization Perspective
• TRIPS: Efficient Vision-and-Language Pre-training with Text-Relevant Image Patch Selection
• Continual Dialogue State Tracking via Example-Guided Question Answering
• A Double Machine Learning Approach to Combining Experimental and Observational Data
• Symbolic Imitation Learning: From Black-Box to Explainable Driving Policies
• Ocassionally Secure: A Comparative Analysis of Code Generation Assistants
• BlockFUL: Enabling Unlearning in Blockchained Federated Learning
• Federated Learning Resilient to Byzantine Attacks and Data Heterogeneity
• FusionDTI: Fine-grained Binding Discovery with Token-level Fusion for Drug-Target Interaction
• Multi-Head RAG: Solving Multi-Aspect Problems with LLMs
• SciRIFF: A Resource to Enhance Language Model Instruction-Following over Scientific Literature
• A Comprehensive Graph Pooling Benchmark: Effectiveness, Robustness and Generalizability
• Position: Towards Bidirectional Human-AI Alignment
• Understanding Transformer Architecture through Continuous Dynamics: A Partial Differential Equation Perspective
• Robot Navigation with Entity-Based Collision Avoidance using Deep Reinforcement Learning
• LLM-3D Print: Large Language Models To Monitor and Control 3D Printing
• A GREAT Architecture for Edge-Based Graph Problems Like TSP
• Parse Trees Guided LLM Prompt Compression
• Distributed AI Platform for the 6G RAN
• Disentangling Regional Primitives for Image Generation
• Extracting Moore Machines from Transformers using Queries and Counterexamples
• Deeper Insights into Deep Graph Convolutional Networks: Stability and Generalization
• NextLocLLM: Location Semantics Modeling and Coordinate-Based Next Location Prediction with LLMs
• Gradient-Free Training of Quantized Neural Networks
• DM-Codec: Distilling Multimodal Representations for Speech Tokenization
• Self-Normalized Resets for Plasticity in Continual Learning
• PACER: Physics Informed Uncertainty Aware Climate Emulator
• When Speculation Spills Secrets: Side Channels via Speculative Decoding In LLMs
• UniTraj: Learning a Universal Trajectory Foundation Model from Billion-Scale Worldwide Traces
• UniHR: Hierarchical Representation Learning for Unified Knowledge Graph Link Prediction
• Integrating Various Software Artifacts for Better LLM-based Bug Localization and Program Repair
• Adapting Chat Language Models Using Only Target Unlabeled Language Data
• Order Matters! An Empirical Study on Large Language Models' Input Order Bias in Software Fault Localization
• Reversed in Time: A Novel Temporal-Emphasized Benchmark for Cross-Modal Video-Text Retrieval
• A Partition Cover Approach to Tokenization
• CGI: Identifying Conditional Generative Models with Example Images
• Mind the Value-Action Gap: Do LLMs Act in Alignment with Their Values?
• FuzzyLight: A Robust Two-Stage Fuzzy Approach for Traffic Signal Control Works in Real Cities
• Principal Components for Neural Network Initialization
• Beyond checkmate: exploring the creative chokepoints in AI text
• Vintix: Action Model via In-Context Reinforcement Learning
• Safety Mirage: How Spurious Correlations Undermine VLM Safety Fine-Tuning and Can Be Mitigated by Machine Unlearning
• Do Larger Language Models Generalize Better? A Scaling Law for Implicit Reasoning at Pretraining Time
• Signal in the Noise: Polysemantic Interference Transfers and Predicts Cross-Model Influence
• FRABench and UFEval: Unified Fine-grained Evaluation with Task and Aspect Generalization
• SPhyR: Spatial-Physical Reasoning Benchmark on Material Distribution
• No Black Boxes: Interpretable and Interactable Predictive Healthcare with Knowledge-Enhanced Agentic Causal Discovery
• Fuzzy Information Evolution with Three-Way Decision in Social Network Group Decision-Making
• DEL-ToM: Inference-Time Scaling for Theory-of-Mind Reasoning via Dynamic Epistemic Logic
• TabularGSM: Understanding the Limitations of LLMs in Tabular Math Reasoning
• HS-STaR: Hierarchical Sampling for Self-Taught Reasoners via Difficulty Estimation and Budget Reallocation
• MineAnyBuild: Benchmarking Spatial Planning for Open-world AI Agents
• Truly Assessing Fluid Intelligence of Large Language Models through Dynamic Reasoning Evaluation
• Orak: A Foundational Benchmark for Training and Evaluating LLM Agents on Diverse Video Games
• VisioMath: Benchmarking Figure-based Mathematical Reasoning in LMMs
• One Patient, Many Contexts: Scaling Medical AI with Contextual Intelligence
• Efficient LLM Collaboration via Planning
• Tiered Agentic Oversight: A Hierarchical Multi-Agent System for Healthcare Safety
• The 4th Dimension for Scaling Model Size
• Breaking Rank Bottlenecks in Knowledge Graph Embeddings
• Bridging Ethical Principles and Algorithmic Methods: An Alternative Approach for Assessing Trustworthiness in AI Systems
• GTA1: GUI Test-time Scaling Agent
• Comprehension Without Competence: Architectural Limits of LLMs in Symbolic Computation and Reasoning
• DuetGraph: Coarse-to-Fine Knowledge Graph Reasoning with Dual-Pathway Global-Local Fusion
• Hierarchical Task Environments as the Next Frontier for Embodied World Models in Robot Soccer
• Neuromorphic Intelligence
• Imagined Autocurricula
• Memory-QA: Answering Recall Questions Based on Multimodal Memories
• LogReasoner: Empowering LLMs with Expert-like Coarse-to-Fine Reasoning for Automated Log Analysis
• The Thinking Spectrum: An Empirical Study of Tunable Reasoning in LLMs through Model Merging
• WordAlchemy: A transformer-based Reverse Dictionary
• Scaling Generalist Data-Analytic Agents
• jina-reranker-v3: Last but Not Late Interaction for Document Reranking
• Scaling with Collapse: Efficient and Predictable Training of LLM Families
• ORPO-Distill: Mixed-Policy Preference Optimization for Cross-Architecture LLM Distillation
• Towards Personalized Deep Research: Benchmarks and Evaluations
• Score Distillation of Flow Matching Models
• MGM-Omni: Scaling Omni LLMs to Personalized Long-Horizon Speech
• Rethinking Entropy Regularization in Large Reasoning Models
• Paired by the Teacher: Turning Unpaired Data into High-Fidelity Pairs for Low-Resource Text Generation
• Fast Feature Field ($\text{F}^3$): A Predictive Representation of Events
• Pretraining Large Language Models with NVFP4
• Chance-constrained Flow Matching for High-Fidelity Constraint-aware Generation
• GSM8K-V: Can Vision Language Models Solve Grade School Math Word Problems in Visual Contexts
• GLASS Flows: Transition Sampling for Alignment of Flow and Diffusion Models
• XQC: Well-conditioned Optimization Accelerates Deep Reinforcement Learning
• EasySteer: A Unified Framework for High-Performance and Extensible LLM Steering
• GHOST: Hallucination-Inducing Image Generation for Multimodal LLMs
• NAIPv2: Debiased Pairwise Learning for Efficient Paper Quality Estimation
• DC-Gen: Post-Training Diffusion Acceleration with Deeply Compressed Latent Space
• DC-VideoGen: Efficient Video Generation with Deep Compression Video Autoencoder
• Incentive-Aligned Multi-Source LLM Summaries
• Guided Diffusion for the Discovery of New Superconductors
• InfoAgent: Advancing Autonomous Information-Seeking Agents
• Query2Triple: Unified Query Encoding for Answering Diverse Complex Queries over Knowledge Graphs
• Taking control: Policies to address extinction risks from advanced AI
• Understanding the Effects of Miscalibrated AI Confidence on User Trust, Reliance, and Decision Efficacy
• Towards Efficient LLM Grounding for Embodied Multi-Agent Collaboration
• VCSearch: Bridging the Gap Between Well-Defined and Ill-Defined Problems in Mathematical Reasoning
• Thread: A Logic-Based Data Organization Paradigm for How-To Question Answering with Retrieval Augmented Generation
• A Voter-Based Stochastic Rejection-Method Framework for Asymptotically Safe Language Model Outputs
• Autonomous Vehicle Controllers From End-to-End Differentiable Simulation
• Neuro-Symbolic Entity Alignment via Variational Inference
• A Neurosymbolic Fast and Slow Architecture for Graph Coloring
• From An LLM Swarm To A PDDL-Empowered HIVE: Planning Self-Executed Instructions In A Multi-Modal Jungle
• GUI Agents: A Survey
• Collaborative Gym: A Framework for Enabling and Evaluating Human-Agent Collaboration
• Broadening Ontologization Design: Embracing Data Pipeline Strategies
• Enabling AI Scientists to Recognize Innovation: A Domain-Agnostic Algorithm for Assessing Novelty
• Visualizing Thought: Conceptual Diagrams Enable Robust Planning in LMMs
• Of-SemWat: High-payload text embedding for semantic watermarking of AI-generated images with arbitrary size
• Putnam-like dataset summary: LLMs as mathematical competition contestants
• Evaluating SAP Joule for Code Generation
• SemShareKV: Efficient KVCache Sharing for Semantically Similar Prompts via Token-Level LSH Matching
• Hierarchical Error Correction for Large Language Models: A Systematic Framework for Domain-Specific AI Quality Enhancement
• Metaphor identification using large language models: A comparison of RAG, prompt engineering, and fine-tuning
• Retro*: Optimizing LLMs for Reasoning-Intensive Document Retrieval
• Uncertainty-Guided Expert-AI Collaboration for Efficient Soil Horizon Annotation
• Vehicle Classification under Extreme Imbalance: A Comparative Study of Ensemble Learning and CNNs
• Scaling Laws and Spectra of Shallow Neural Networks in the Feature Learning Regime
• OpenGPT-4o-Image: A Comprehensive Dataset for Advanced Image Generation and Editing
• Segmentor-Guided Counterfactual Fine-Tuning for Image Synthesis
• When Greedy Wins: Emergent Exploitation Bias in Meta-Bandit LLM Training
• Scalable GANs with Transformers
• MobileLLM-R1: Exploring the Limits of Sub-Billion Language Model Reasoners with Open Training Recipes
• Learning Distinguishable Representations in Deep Q-Networks for Linear Transfer
• MSG: Multi-Stream Generative Policies for Sample-Efficient Robotic Manipulation
• SecInfer: Preventing Prompt Injection via Inference-time Scaling
• Random Policy Valuation is Enough for LLM Reasoning with Verifiable Rewards
• Light-SQ: Structure-aware Shape Abstraction with Superquadrics for Generated Meshes
• Generalized Correctness Models: Learning Calibrated and Model-Agnostic Correctness Predictors from Historical Patterns
• CLASP: Adaptive Spectral Clustering for Unsupervised Per-Image Segmentation
• AIRoA MoMa Dataset: A Large-Scale Hierarchical Dataset for Mobile Manipulation
• Ultra-Fast Language Generation via Discrete Diffusion Divergence Instruct
• Fast Real-Time Pipeline for Robust Arm Gesture Recognition
• Large Language Models for Software Testing: A Research Roadmap
• Hyperdimensional Probe: Decoding LLM Representations via Vector Symbolic Architectures
• Optimizing Privacy-Preserving Primitives to Support LLM-Scale Applications
• BRIDGE - Building Reinforcement-Learning Depth-to-Image Data Generation Engine for Monocular Depth Estimation
• UniLat3D: Geometry-Appearance Unified Latents for Single-Stage 3D Generation
• CoTune: Co-evolutionary Configuration Tuning
• SANA-Video: Efficient Video Generation with Block Linear Diffusion Transformer
• T-POP: Test-Time Personalization with Online Preference Feedback
• FedPOB: Sample-Efficient Federated Prompt Optimization via Bandits
• Circuit-Aware Reward Training: A Mechanistic Framework for Longtail Robustness in RLHF
• Discrete Variational Autoencoding via Policy Search
• Q-Net: Transferable Queue Length Estimation via Kalman-based Neural Networks
• A TRIANGLE Enables Multimodal Alignment Beyond Cosine Similarity
• Robust Policy Expansion for Offline-to-Online RL under Diverse Data Corruption
• Surjective Independence of Causal Influences for Local Bayesian Network Structures
• VSSFlow: Unifying Video-conditioned Sound and Speech Generation via Joint Learning
• VTPerception-R1: Enhancing Multimodal Reasoning via Explicit Visual and Textual Perceptual Grounding
• Quantifying Generalisation in Imitation Learning
• Sparse Autoencoders Make Audio Foundation Models more Explainable
• Fidelity-Aware Data Composition for Robust Robot Generalization
• Causal-Adapter: Taming Text-to-Image Diffusion for Faithful Counterfactual Generation
• DSAT-HD: Dual-Stream Adaptive Transformer with Hybrid Decomposition for Multivariate Time Series Forecasting
• RDD: Pareto Analysis of the Rate-Distortion-Distinguishability Trade-off
• Intelligent Optimization of Wireless Access Point Deployment for Communication-Based Train Control Systems Using Deep Reinforcement Learning
• CMT: Mid-Training for Efficient Learning of Consistency, Mean Flow, and Flow Map Models
• CORE-3D: Context-aware Open-vocabulary Retrieval by Embeddings in 3D
• Short window attention enables long-term memorization
• Deep Reinforcement Learning in Action: Real-Time Control of Vortex-Induced Vibrations
• AdaThink-Med: Medical Adaptive Thinking with Uncertainty-Guided Length Calibration
• Bandits roaming Hilbert space
• PoseDiff: A Unified Diffusion Model Bridging Robot Pose Estimation and Video-to-Action Control
• Algorithms and data structures for automatic precision estimation of neural networks
• Can you SPLICE it together? A Human Curated Benchmark for Probing Visual Reasoning in VLMs
• Identity Bridge: Enabling Implicit Reasoning via Shared Latent Memory
• VNODE: A Piecewise Continuous Volterra Neural Network
• Community detection robustness of graph neural networks
• InfLLM-V2: Dense-Sparse Switchable Attention for Seamless Short-to-Long Adaptation
• Understanding the Dilemma of Unlearning for Large Language Models
• Reference-Free Rating of LLM Responses via Latent Information
• Data-Driven Discrete Geofence Design Using Binary Quadratic Programming
• LAMP-PRo: Label-aware Attention for Multi-label Prediction of DNA- and RNA-binding Proteins using Protein Language Models
• Cycle Diffusion Model for Counterfactual Image Generation
• Adversarial Reinforcement Learning Framework for ESP Cheater Simulation
• SimuHome: A Temporal- and Environment-Aware Benchmark for Smart Home LLM Agents
• Let LLMs Speak Embedding Languages: Generative Text Embeddings via Iterative Contrastive Refinement
• DiffuGuard: How Intrinsic Safety is Lost and Found in Diffusion Large Language Models
• Q-Mirror: Unlocking the Multi-Modal Potential of Scientific Text-Only QA Pairs
• Bridging the behavior-neural gap: A multimodal AI reveals the brain's geometry of emotion more accurately than human self-reports
• A study of Universal ODE approaches to predicting soil organic carbon
• Dual Mechanisms of Value Expression: Intrinsic vs. Prompted Values in LLMs
• TraitSpaces: Towards Interpretable Visual Creativity for Human-AI Co-Creation
• Towards Generalizable PDE Dynamics Forecasting via Physics-Guided Invariant Learning
• Dynamic Orchestration of Multi-Agent System for Real-World Multi-Image Agricultural VQA
• Beyond Repetition: Text Simplification and Curriculum Learning for Data-Constrained Pretraining
• An Enhanced Pyramid Feature Network Based on Long-Range Dependencies for Multi-Organ Medical Image Segmentation
• UI-UG: A Unified MLLM for UI Understanding and Generation
• Uni-X: Mitigating Modality Conflict with a Two-End-Separated Architecture for Unified Multimodal Models
• Watermarking Diffusion Language Models
• From Satellite to Street: A Hybrid Framework Integrating Stable Diffusion and PanoGAN for Consistent Cross-View Synthesis
• Evolution Strategies at Scale: LLM Fine-Tuning Beyond Reinforcement Learning
• REALIGN: Regularized Procedure Alignment with Matching Video Embeddings via Partial Gromov-Wasserstein Optimal Transport
• HarmMetric Eval: Benchmarking Metrics and Judges for LLM Harmfulness Assessment
• Vid-LLM: A Compact Video-based 3D Multimodal LLM with Reconstruction-Reasoning Synergy
• LLaDA-MoE: A Sparse MoE Diffusion Language Model
• The 2025 OpenAI Preparedness Framework does not guarantee any AI risk mitigation practices: a proof-of-concept for affordance analyses of AI safety policies
• Multilingual Text-to-SQL: Benchmarking the Limits of Language Models with Collaborative Language Agents
• Hybrid Layer-Wise ANN-SNN With Surrogate Spike Encoding-Decoding Structure
• ScatterAD: Temporal-Topological Scattering Mechanism for Time Series Anomaly Detection
• CLQ: Cross-Layer Guided Orthogonal-based Quantization for Diffusion Transformers
• A Data-Centric Perspective on the Influence of Image Data Quality in Machine Learning Models
• Multi-Item-Query Attention for Stable Sequential Recommendation
• Alternatives To Next Token Prediction In Text Generation - A Survey
• EOE: Evolutionary Optimization of Experts for Training Language Models
• An Agent-Based Framework for Automated Higher-Voice Harmony Generation
• Moravec's Paradox and Restrepo's Model: Limits of AGI Automation in Growth
• LaMoGen: Laban Movement-Guided Diffusion for Text-to-Motion Generation
• Euclid's Gift: Enhancing Spatial Perception and Reasoning in Vision-Language Models via Geometric Surrogate Tasks
• Mitigating Visual Hallucinations via Semantic Curriculum Preference Optimization in MLLMs
• LLM DNA: Tracing Model Evolution via Functional Representations
• Specialization after Generalization: Towards Understanding Test-Time Training in Foundation Models
• Agentic Specification Generator for Move Programs
• PhysiAgent: An Embodied Agent Framework in Physical World
• SLA: Beyond Sparsity in Diffusion Transformers via Fine-Tunable Sparse-Linear Attention
• FrameMind: Frame-Interleaved Chain-of-Thought for Video Reasoning via Reinforcement Learning
• From Edge to HPC: Investigating Cross-Facility Data Streaming Architectures
• GPS-MTM: Capturing Pattern of Normalcy in GPS-Trajectories with self-supervised learning
• End-to-end Topographic Auditory Models Replicate Signatures of Human Auditory Cortex
• PartnerMAS: An LLM Hierarchical Multi-Agent Framework for Business Partner Selection on High-Dimensional Features
• A Second-Order Perspective on Pruning at Initialization and Knowledge Transfer
• In-Context Compositional Q-Learning for Offline Reinforcement Learning
• A Small Math Model: Recasting Strategy Choice Theory in an LLM-Inspired Architecture
• AQUAIR: A High-Resolution Indoor Environmental Quality Dataset for Smart Aquaculture Monitoring
• Uncovering Grounding IDs: How External Cues Shape Multi-Modal Binding
• PEARL: Peer-Enhanced Adaptive Radio via On-Device LLM
• Large-Scale Constraint Generation - Can LLMs Parse Hundreds of Constraints?
• PerfBench: Can Agents Resolve Real-World Performance Bugs?
• GEAR: A General Evaluation Framework for Abductive Reasoning
• Ancestry Tree Clustering for Particle Filter Diversity Maintenance
• The Impossibility of Inverse Permutation Learning in Transformer Models
• BOSfM: A View Planning Framework for Optimal 3D Reconstruction of Agricultural Scenes
• ASTROCO: Self-Supervised Conformer-Style Transformers for Light-Curve Embeddings
• EYE-DEX: Eye Disease Detection and EXplanation System
• Your thoughts tell who you are: Characterize the reasoning patterns of LRMs
• TENET: Leveraging Tests Beyond Validation for Code Generation
• Accelerating Cerebral Diagnostics with BrainFusion: A Comprehensive MRI Tumor Framework
• Memory Transfer Planning: LLM-driven Context-Aware Code Adaptation for Robot Manipulation
• LatXGen: Towards Radiation-Free and Accurate Quantitative Analysis of Sagittal Spinal Alignment Via Cross-Modal Radiographic View Synthesis
• Stable Forgetting: Bounded Parameter-Efficient Unlearning in LLMs
• Retrieval-augmented GUI Agents with Generative Guidelines
• Beyond Overall Accuracy: A Psychometric Deep Dive into the Topic-Specific Medical Capabilities of 80 Large Language Models
• Talk in Pieces, See in Whole: Disentangling and Hierarchical Aggregating Representations for Language-based Object Detection
• AceSearcher: Bootstrapping Reasoning and Search for LLMs via Reinforced Self-Play
• Chat to Chip: Large Language Model Based Design of Arbitrarily Shaped Metasurfaces
• Can Large Language Models Express Uncertainty Like Human?
• Group-Relative REINFORCE Is Secretly an Off-Policy Algorithm: Demystifying Some Myths About GRPO and Its Friends
• BALR-SAM: Boundary-Aware Low-Rank Adaptation of SAM for Resource-Efficient Medical Image Segmentation
• BeyondBench: Benchmark-Free Evaluation of Reasoning in Language Models
• Metamorphic Testing for Audio Content Moderation Software
• Conda: Column-Normalized Adam for Training Large Language Models Faster
• ViReSkill: Vision-Grounded Replanning with Skill Memory for LLM-Based Planning in Lifelong Robot Learning
• Uni-NTFM: A Unified Foundation Model for EEG Signal Representation Learning
• ChessArena: A Chess Testbed for Evaluating Strategic Reasoning Capabilities of Large Language Models
• SafeFlowMatcher: Safe and Fast Planning using Flow Matching with Control Barrier Functions
• Prompt and Parameter Co-Optimization for Large Language Models
• Graph Foundation Models: Bridging Language Model Paradigms and Graph Optimization
• SHAPoint: Task-Agnostic, Efficient, and Interpretable Point-Based Risk Scoring via Shapley Values
• Accuracy-Robustness Trade Off via Spiking Neural Network Gradient Sparsity Trail
• Knowledge-Level Consistency Reinforcement Learning: Dual-Fact Alignment for Long-Form Factuality
• From Personal to Collective: On the Role of Local and Global Memory in LLM Personalization
• Knowledge Homophily in Large Language Models
• Sequence Pathfinder for Multi-Agent Pickup and Delivery in the Warehouse
• GroupCoOp: Group-robust Fine-tuning via Group Prompt Learning
• From Unstable to Playable: Stabilizing Angry Birds Levels via Object Segmentation
• Enhancing LLM Steering through Sparse Autoencoder-Based Vector Refinement
• FedAgentBench: Towards Automating Real-world Federated Medical Image Analysis with Server-Client LLM Agents
• Tequila: Trapping-free Ternary Quantization for Large Language Models
• Navigating the Labyrinth: Path-Sensitive Unit Test Generation with Large Language Models
• IndexNet: Timestamp and Variable-Aware Modeling for Time Series Forecasting
• A Multi-Camera Vision-Based Approach for Fine-Grained Assembly Quality Control
• Space Group Conditional Flow Matching
• HFuzzer: Testing Large Language Models for Package Hallucinations via Phrase-based Fuzzing
• Adversarial Diffusion for Robust Reinforcement Learning
• GSID: Generative Semantic Indexing for E-Commerce Product Understanding
• Efficient Multi-turn RL for GUI Agents via Decoupled Training and Adaptive Data Curation
• Taught Well Learned Ill: Towards Distillation-conditional Backdoor Attack
• Multi-Value-Product Retrieval-Augmented Generation for Industrial Product Attribute Value Identification
• Not All Tokens are Guided Equal: Improving Guidance in Visual Autoregressive Models
• Disentangling Score Content and Performance Style for Joint Piano Rendering and Transcription
• PCRI: Measuring Context Robustness in Multimodal Models for Enterprise Applications
• Tunable-Generalization Diffusion Powered by Self-Supervised Contextual Sub-Data for Low-Dose CT Reconstruction
• Towards Understanding Subliminal Learning: When and How Hidden Biases Transfer
• Gradient Flow Convergence Guarantee for General Neural Network Architectures
• Dynamic Orthogonal Continual Fine-tuning for Mitigating Catastrophic Forgettings
• Preserving Cross-Modal Stability for Visual Unlearning in Multimodal Scenarios
• Interpreting deep learning-based stellar mass estimation via causal analysis and mutual information decomposition
• EWC-Guided Diffusion Replay for Exemplar-Free Continual Learning in Medical Imaging
• Continual Learning to Generalize Forwarding Strategies for Diverse Mobile Wireless Networks
• Graph Mixing Additive Networks
• Taming Masked Diffusion Language Models via Consistency Trajectory Reinforcement Learning with Fewer Decoding Step
• HiViS: Hiding Visual Tokens from the Drafter for Speculative Decoding in Vision-Language Models
• Diffusion Models are Kelly Gamblers
• Easy Turn: Integrating Acoustic and Linguistic Modalities for Robust Turn-Taking in Full-Duplex Spoken Dialogue Systems
• Explore-Execute Chain: Towards an Efficient Structured Reasoning Paradigm
• Vision-Grounded Machine Interpreting: Improving the Translation Process through Visual Cues
• MAD-PINN: A Decentralized Physics-Informed Machine Learning Framework for Safe and Optimal Multi-Agent Control
• Toward Preference-aligned Large Language Models via Residual-based Model Steering
• The Hidden Costs of Translation Accuracy: Distillation, Quantization, and Environmental Impact
• Guide: Generalized-Prior and Data Encoders for DAG Estimation
• The AI Agent Code of Conduct: Automated Guardrail Policy-as-Prompt Synthesis
• MCPMark: A Benchmark for Stress-Testing Realistic and Comprehensive MCP Use
• RAVEN: Resilient Aerial Navigation via Open-Set Semantic Memory and Behavior Adaptation
• Node Classification via Simplicial Interaction with Augmented Maximal Clique Selection
• Benchmarking LLM-Assisted Blue Teaming via Standardized Threat Hunting
• Uncovering Vulnerabilities of LLM-Assisted Cyber Threat Intelligence
• Towards Efficient CoT Distillation: Self-Guided Rationale Selector for Better Performance with Fewer Rationales
• ML-Asset Management: Curation, Discovery, and Utilization
• Improving the Efficiency of LLM Agent Systems through Trajectory Reduction
• Toward a Holistic Approach to Continual Model Merging
• Timber: Training-free Instruct Model Refining with Base via Effective Rank
• Multi-Level Heterogeneous Knowledge Transfer Network on Forward Scattering Center Model for Limited Samples SAR ATR
• Characteristic Root Analysis and Regularization for Linear Time Series Forecasting
• InteractMove: Text-Controlled Human-Object Interaction Generation in 3D Scenes with Movable Objects
• GraphIFE: Rethinking Graph Imbalance Node Classification via Invariant Learning
• BioVessel-Net and RetinaMix: Unsupervised Retinal Vessel Segmentation from OCTA Images
• Generalizable Speech Deepfake Detection via Information Bottleneck Enhanced Adversarial Alignment
• RIV: Recursive Introspection Mask Diffusion Vision Language Model
• LightFair: Towards an Efficient Alternative for Fair T2I Diffusion via Debiasing Pre-trained Text Encoders
• ReWatch-R1: Boosting Complex Video Reasoning in Large Vision-Language Models through Agentic Data Synthesis
• Focusing on What Matters: Object-Agent-centric Tokenization for Vision Language Action models
• Aligning LLMs for Multilingual Consistency in Enterprise Applications
• Pure Node Selection for Imbalanced Graph Node Classification
• Calibration Meets Reality: Making Machine Learning Predictions Trustworthy
• Beyond Greedy Exits: Improved Early Exit Decisions for Risk Control and Reliability
• Graph Neural Networks with Diversity-aware Neighbor Selection and Dynamic Multi-scale Fusion for Multivariate Time Series Forecasting
• RCI: A Score for Evaluating Global and Local Reasoning in Multimodal Benchmarks
• Towards a Comprehensive Scaling Law of Mixture-of-Experts
• Joint Hybrid Beamforming and Artificial Noise Design for Secure Multi-UAV ISAC Networks
• Estimating Time Series Foundation Model Transferability via In-Context Learning
• CrimEdit: Controllable Editing for Counterfactual Object Removal, Insertion, and Movement
• Bridging Discrete and Continuous RL: Stable Deterministic Policy Gradient with Martingale Characterization
• AdaPtis: Reducing Pipeline Bubbles with Adaptive Pipeline Parallelism on Heterogeneous Models
• Video Panels for Long Video Understanding
• AudioMoG: Guiding Audio Generation with Mixture-of-Guidance
• M3DLayout: A Multi-Source Dataset of 3D Indoor Layouts and Structured Descriptions for 3D Generation
• LUQ: Layerwise Ultra-Low Bit Quantization for Multimodal Large Language Models
• HieraTok: Multi-Scale Visual Tokenizer Improves Image Reconstruction and Generation
• Compose and Fuse: Revisiting the Foundational Bottlenecks in Multimodal Reasoning
• LocoFormer: Generalist Locomotion via Long-context Adaptation
• Poivre: Self-Refining Visual Pointing with Reinforcement Learning
• PVTAdpNet: Polyp Segmentation using Pyramid vision transformer with a novel Adapter block
• Understanding Textual Capability Degradation in Speech LLMs via Parameter Importance Analysis
• PARROT: A Benchmark for Evaluating LLMs in Cross-System SQL Translation
• DentVLM: A Multimodal Vision-Language Model for Comprehensive Dental Diagnosis and Enhanced Clinical Practice
• ABC-Eval: Benchmarking Large Language Models on Symbolic Music Understanding and Instruction Following
• Dynamic-TreeRPO: Breaking the Independent Trajectory Bottleneck with Structured Sampling
• Dual-Space Smoothness for Robust and Balanced LLM Unlearning
• AI Education in Higher Education: A Taxonomy for Curriculum Reform and the Mission of Knowledge
• MedCritical: Enhancing Medical Reasoning in Small Language Models via Self-Collaborative Correction
• Alignment through Meta-Weighted Online Sampling: Bridging the Gap between Data Generation and Preference Optimization
• Graph Your Own Prompt
• CCD: Mitigating Hallucinations in Radiology MLLMs via Clinical Contrastive Decoding
• Train Once, Answer All: Many Pretraining Experiments for the Cost of One
• Enhanced Fracture Diagnosis Based on Critical Regional and Scale Aware in YOLO
• PATCH: Learnable Tile-level Hybrid Sparsity for LLMs
• Hybrid Graph Embeddings and Louvain Algorithm for Unsupervised Community Detection
• Retrieval-Constrained Decoding Reveals Underestimated Parametric Knowledge in Language Models
• Enhancing Communication Efficiency in FL with Adaptive Gradient Quantization and Communication Frequency Optimization
• NeuroBridge: Using Generative AI to Bridge Cross-neurotype Communication Differences through Neurotypical Perspective-taking
• AudioRole: An Audio Dataset for Character Role-Playing in Large Language Models
• S$^3$F-Net: A Multi-Modal Approach to Medical Image Classification via Spatial-Spectral Summarizer Fusion Network
• Factor Decorrelation Enhanced Data Removal from Deep Predictive Models
• AudioFuse: Unified Spectral-Temporal Learning via a Hybrid ViT-1D CNN Architecture for Robust Phonocardiogram Classification
• Data-Efficient Training by Evolved Sampling
• Generative Evolutionary Meta-Solver (GEMS): Scalable Surrogate-Free Multi-Agent Learning
• Multi-Modal Manipulation via Multi-Modal Policy Consensus
• Memory-Efficient Fine-Tuning via Low-Rank Activation Compression
• Text-Based Approaches to Item Difficulty Modeling in Large-Scale Assessments: A Systematic Review
• Revisiting Multivariate Time Series Forecasting with Missing Values
• The Impact of Role Design in In-Context Learning for Large Language Models
• Enhancing Polyp Segmentation via Encoder Attention and Dynamic Kernel Update
• From Human Annotation to Automation: LLM-in-the-Loop Active Learning for Arabic Sentiment Analysis
• Evaluating point-light biological motion in multimodal large language models
• ReliabilityRAG: Effective and Provably Robust Defense for RAG-based Web-Search
• Privy: Envisioning and Mitigating Privacy Risks for Consumer-facing AI Product Concepts
• Imaging-Based Mortality Prediction in Patients with Systemic Sclerosis
• On the Shelf Life of Fine-Tuned LLM Judges: Future Proofing, Backward Compatibility, and Question Generalization
• End-to-End Deep Learning for Predicting Metric Space-Valued Outputs
• Disentanglement of Variations with Multimodal Generative Modeling
• Automatic Speech Recognition for Greek Medical Dictation
• Fusing Sequence Motifs and Pan-Genomic Features: Antimicrobial Resistance Prediction using an Explainable Lightweight 1D CNN-XGBoost Ensemble
• Pancreas Part Segmentation under Federated Learning Paradigm
• HTMA-Net: Towards Multiplication-Avoiding Neural Networks via Hadamard Transform and In-Memory Computing
• Open-Vocabulary Spatio-Temporal Scene Graph for Robot Perception and Teleoperation Planning
• Liaohe-CobotMagic-PnP: an Imitation Learning Dataset of Intelligent Robot for Industrial Applications
• RHYTHM: Reasoning with Hierarchical Temporal Tokenization for Human Mobility
• C$^2$GSPG: Confidence-calibrated Group Sequence Policy Gradient towards Self-aware Reasoning
• Trust Region Reward Optimization and Proximal Inverse Reward Optimization Algorithm
• Deep Learning-Based Detection of Cognitive Impairment from Passive Smartphone Sensing with Routine-Aware Augmentation and Demographic Personalization
• Dense associative memory on the Bures-Wasserstein space
• TRAX: TRacking Axles for Accurate Axle Count Estimation
• WARBERT: A Hierarchical BERT-based Model for Web API Recommendation
• PARL-MT: Learning to Call Functions in Multi-Turn Conversation with Progress Awareness
• Towards Monotonic Improvement in In-Context Reinforcement Learning
• One-Shot Multi-Label Causal Discovery in High-Dimensional Event Sequences
• Leave No Observation Behind: Real-time Correction for VLA Action Chunks
• SPEC-RL: Accelerating On-Policy Reinforcement Learning via Speculative Rollouts
• Patch Rebirth: Toward Fast and Transferable Model Inversion of Vision Transformers
• Self-Consistency as a Free Lunch: Reducing Hallucinations in Vision-Language Models via Self-Reflection
• Online Dynamic Goal Recognition in Gym Environments
• Adaptive Token-Weighted Differential Privacy for LLMs: Not All Tokens Require Equal Protection
• Learning Regional Monsoon Patterns with a Multimodal Attention U-Net
• Vid-Freeze: Protecting Images from Malicious Image-to-Video Generation via Temporal Freezing
• Continuous-Time Reinforcement Learning for Asset-Liability Management
• A2D: Any-Order, Any-Step Safety Alignment for Diffusion Language Models
• A Neural ODE Approach to Aircraft Flight Dynamics Modelling
• Seeing Symbols, Missing Cultures: Probing Vision-Language Models' Reasoning on Fire Imagery and Cultural Meaning
• MELCOT: A Hybrid Learning Architecture with Marginal Preservation for Matrix-Valued Regression
• Scaling LLM Test-Time Compute with Mobile NPU on Smartphones
• Robust Fine-Tuning from Non-Robust Pretrained Models: Mitigating Suboptimal Transfer With Adversarial Scheduling
• Space Robotics Bench: Robot Learning Beyond Earth
• Beyond Model Ranking: Predictability-Aligned Evaluation for Time Series Forecasting
• Signal Preserving Weight Initialization for Odd-Sigmoid Activations
• Causally-Enhanced Reinforcement Policy Optimization
• CoPatch: Zero-Shot Referring Image Segmentation by Leveraging Untapped Spatial Knowledge in CLIP
• Towards Quantum-Ready Blockchain Fraud Detection via Ensemble Graph Neural Networks
• Boundary on the Table: Efficient Black-Box Decision-Based Attacks for Structured Data
• Adaptive Margin RLHF via Preference over Preferences
• Patient-specific Biomolecular Instruction Tuning
• Observation-Free Attacks on Online Learning to Rank
• Scalable Wi-Fi RSS-Based Indoor Localization via Automatic Vision-Assisted Calibration
• From Noise to Knowledge: A Comparative Study of Acoustic Anomaly Detection Models in Pumped-storage Hydropower Plants
• Convolutional Set Transformer
• Extract-0: A Specialized Language Model for Document Information Extraction
• TY-RIST: Tactical YOLO Tricks for Real-time Infrared Small Target Detection
• Rethinking Large Language Model Distillation: A Constrained Markov Decision Process Perspective
• Soft-Di[M]O: Improving One-Step Discrete Image Generation with Soft Embeddings
• Large language models management of medications: three performance analyses
• MonoCon: A general framework for learning ultra-compact high-fidelity representations using monotonicity constraints
• Compute-Optimal Quantization-Aware Training
• Unsupervised Speech Enhancement using Data-defined Priors
• What Matters More For In-Context Learning under Matched Compute Budgets: Pretraining on Natural Text or Incorporating Targeted Synthetic Examples?
• Tiny-QMoE
• Functional Critic Modeling for Provably Convergent Off-Policy Actor-Critic
• ADAM: A Diverse Archive of Mankind for Evaluating and Enhancing LLMs in Biographical Reasoning
• Physically Plausible Multi-System Trajectory Generation and Symmetry Discovery
• MoE-PHDS: One MoE checkpoint for flexible runtime sparsity
• LLM Watermark Evasion via Bias Inversion
• Tracing the Representation Geometry of Language Models from Pretraining to Post-training
• DPFNAS: Differential Privacy-Enhanced Federated Neural Architecture Search for 6G Edge Intelligence
• Sensor-Adaptive Flood Mapping with Pre-trained Multi-Modal Transformers across SAR and Multispectral Modalities
• GeLoc3r: Enhancing Relative Camera Pose Regression with Geometric Consistency Regularization
• Look Back to Reason Forward: Revisitable Memory for Long-Context LLM Agents
• Virus Infection Attack on LLMs: Your Poisoning Can Spread "VIA" Synthetic Data
• IsingFormer: Augmenting Parallel Tempering With Learned Proposals
• MMeViT: Multi-Modal ensemble ViT for Post-Stroke Rehabilitation Action Recognition
• Beyond Aggregation: Guiding Clients in Heterogeneous Federated Learning
• Understanding Language Prior of LVLMs by Contrasting Chain-of-Embedding
• Local Success Does Not Compose: Benchmarking Large Language Models for Compositional Formal Verification
• Semantic Voting: A Self-Evaluation-Free Approach for Efficient LLM Self-Improvement on Unverifiable Open-ended Tasks
• From Evidence to Trajectory: Abductive Reasoning Path Synthesis for Training Retrieval-Augmented Generation Agents
• Localizing Adversarial Attacks To Produces More Imperceptible Noise
• TRUEBench: Can LLM Response Meet Real-world Constraints as Productivity Assistant?
• IBiT: Utilizing Inductive Biases to Create a More Data Efficient Attention Mechanism
• LayoutAgent: A Vision-Language Agent Guided Compositional Diffusion for Spatial Layout Planning
• A Data-Driven Framework for Digital Transformation in Smart Cities: Integrating AI, Dashboards, and IoT Readiness
• A Meta-Analysis of LLM Effects on Students across Qualification, Socialisation, and Subjectification
• Prompt-aware classifier free guidance for diffusion models
• Multi-Modal Sentiment Analysis with Dynamic Attention Fusion
• Bidirectional Intention Inference Enhances LLMs' Defense Against Multi-Turn Jailbreak Attacks
• Rebuild AC Power Flow Models with Graph Attention Networks
• Automated Formative Feedback for Short-form Writing: An LLM-Driven Approach and Adoption Analysis
• Regulating the Agency of LLM-based Agents
• Consistency Models as Plug-and-Play Priors for Inverse Problems
• CompareBench: A Benchmark for Visual Comparison Reasoning in Vision-Language Models
• Painless Activation Steering: An Automated, Lightweight Approach for Post-Training Large Language Models
• Learning What To Hear: Boosting Sound-Source Association For Robust Audiovisual Instance Segmentation
• Societal Capacity Assessment Framework: Measuring Resilience to Inform Advanced AI Risk Management
• Index-MSR: A high-efficiency multimodal fusion framework for speech recognition
• Defending MoE LLMs against Harmful Fine-Tuning via Safety Routing Alignment
• MIRAGE: Multi-hop Reasoning with Ambiguity Evaluation for Illusory Questions
• Variance-Bounded Evaluation without Ground Truth: VB-Score
• Self-driving cars: Are we there yet?
• Persistent Autoregressive Mapping with Traffic Rules for Autonomous Driving
• Red Teaming Quantum-Resistant Cryptographic Standards: A Penetration Testing Framework Integrating AI and Quantum Security
• MILR: Improving Multimodal Image Generation via Test-Time Latent Reasoning
• UESA-Net: U-Shaped Embedded Multidirectional Shrinkage Attention Network for Ultrasound Nodule Segmentation
• In-Context Learning can Perform Continual Learning Like Humans
• A theoretical guarantee for SyncRank
• Differentially Private Two-Stage Gradient Descent for Instrumental Variable Regression
• Generative Modeling and Decision Fusion for Unknown Event Detection and Classification Using Synchrophasor Data
• VideoScore2: Think before You Score in Generative Video Evaluation
• MTRec: Learning to Align with User Preferences via Mental Reward Models
• MMPB: It's Time for Multi-Modal Personalization
• Dynamic Buffers: Cost-Efficient Planning for Tabletop Rearrangement with Stacking
• Efficient Fine-Grained GPU Performance Modeling for Distributed Deep Learning of LLM
• Bridging Language Models and Formal Methods for Intent-Driven Optical Network Design
• Seeing Isn't Believing: Context-Aware Adversarial Patch Synthesis via Conditional GAN
• Multimodal Slice Interaction Network Enhanced by Transfer Learning for Precise Segmentation of Internal Gross Tumor Volume in Lung Cancer PET/CT Imaging
• On the Self-awareness of Large Reasoning Models' Capability Boundaries
• Spatial-Functional awareness Transformer-based graph archetype contrastive learning for Decoding Visual Neural Representations from EEG
• From Ambiguity to Verdict: A Semiotic-Grounded Multi-Perspective Agent for LLM Logical Reasoning
• TimeOmni-1: Incentivizing Complex Reasoning with Time Series in Large Language Models
• Query Circuits: Explaining How Language Models Answer User Prompts
• Pushing LLMs to Their Logical Reasoning Bound: The Role of Data Reasoning Intensity
• PhysicsMinions: Winning Gold Medals in the Latest Physics Olympiads with a Coevolutionary Multimodal Multi-Agent System
• The Emergence of Social Science of Large Language Models
• RealUnify: Do Unified Models Truly Benefit from Unification? A Comprehensive Benchmark
• Neural network embeddings recover value dimensions from psychometric survey items on par with human data
• Meta-Learning Theory-Informed Inductive Biases using Deep Kernel Gaussian Processes
• MASLegalBench: Benchmarking Multi-Agent Systems in Deductive Legal Reasoning
• When Autonomous Vehicle Meets V2X Cooperative Perception: How Far Are We?
• KIRETT - A wearable device to support rescue operations using artificial intelligence to improve first aid
• Agentic Exploration of Physics Models
• CLPO: Curriculum Learning meets Policy Optimization for LLM Reasoning
• Scaling Synthetic Task Generation for Agents via Exploration
• Cogito, Ergo Ludo: An Agent that Learns to Play by Reasoning and Planning
• HeDA: An Intelligent Agent System for Heatwave Risk Discovery through Automated Knowledge Graph Construction and Multi-layer Risk Propagation Analysis
• From $f(x)$ and $g(x)$ to $f(g(x))$: LLMs Learn New Skills in RL by Composing Old Ones
• The Era of Real-World Human Interaction: RL from User Conversations
• Vision-and-Language Navigation with Analogical Textual Descriptions in LLMs
• ReasoningBank: Scaling Agent Self-Evolving with Reasoning Memory
• Visual serial processing deficits explain divergences in human and VLM reasoning
• UniAPL: A Unified Adversarial Preference Learning Framework for Instruct-Following
• Who's Your Judge? On the Detectability of LLM-Generated Judgments
• Prosody-Adaptable Audio Codecs for Zero-Shot Voice Conversion via In-Context Learning
• BenLOC: A Benchmark for Learning to Configure MIP Optimizers
• YOLO-based Bearing Fault Diagnosis With Continuous Wavelet Transform
• Agentic DDQN-Based Scheduling for Licensed and Unlicensed Band Allocation in Sidelink Networks
• Green Learning for STAR-RIS mmWave Systems with Implicit CSI
• How are Scientific Concepts Birthed? Typing Rules of Concept Formation in Theoretical Physics Reasoning
• Sustainable LSTM-Based Precoding for RIS-Aided mmWave MIMO Systems with Implicit CSI
• GOAT: A Large Dataset of Paired Guitar Audio Recordings and Tablatures
• How good are LLMs at Retrieving Documents in a Specific Domain?
• Fairness for niche users and providers: algorithmic choice and profile portability
• Next Point-of-interest (POI) Recommendation Model Based on Multi-modal Spatio-temporal Context Feature Embedding
• PISA: An AI Pipeline for Interpretable-by-design Survival Analysis Providing Multiple Complexity-Accuracy Trade-off Models
• Learning Hyperspectral Images with Curated Text Prompts for Efficient Multimodal Alignment
• Advancing Audio-Visual Navigation Through Multi-Agent Collaboration in 3D Environments
• Enhancing Cluster Scheduling in HPC: A Continuous Transfer Learning for Real-Time Optimization
• AccessEval: Benchmarking Disability Bias in Large Language Models
• Intelligent Load Balancing in Cloud Computer Systems
• GZSL-MoE: Apprentissage G{\'e}n{\'e}ralis{\'e} Z{\'e}ro-Shot bas{\'e} sur le M{\'e}lange d'Experts pour la Segmentation S{\'e}mantique de Nuages de Points 3DAppliqu{\'e} {\`a} un Jeu de Donn{\'e}es d'Environnement de Collaboration Humain-Robot
• AnveshanaAI: A Multimodal Platform for Adaptive AI/ML Education through Automated Question Generation and Interactive Assessment
• Mix-Ecom: Towards Mixed-Type E-Commerce Dialogues with Complex Domain Rules
• AgentGuard: Runtime Verification of AI Agents
• Rethinking Reward Miscalibration of GRPO in Agentic RL
• Quant Fever, Reasoning Blackholes, Schrodinger's Compliance, and More: Probing GPT-OSS-20B
• From Neural Networks to Logical Theories: The Correspondence between Fibring Modal Logics and Fibring Neural Networks
• Conditional Advantage Estimation for Reinforcement Learning in Large Reasoning Models
• Automatic selection of primary studies in systematic reviews with evolutionary rule-based classification
• TusoAI: Agentic Optimization for Scientific Methods
• LLM/Agent-as-Data-Analyst: A Survey
• Future-Proofing Programmers: Optimal Knowledge Tracing for AI-Assisted Personalized Education
• Do Repetitions Matter? Strengthening Reliability in LLM Evaluations
• Fathom-DeepResearch: Unlocking Long Horizon Information Retrieval and Synthesis for SLMs
• Transparent, Evaluable, and Accessible Data Agents: A Proof-of-Concept Framework
• Reasoning or Retrieval? A Study of Answer Attribution on Large Reasoning Models
• Robust Preference Optimization: Aligning Language Models with Noisy Preference Feedback
• Humanline: Online Alignment as Perceptual Loss
• ELHPlan: Efficient Long-Horizon Task Planning for Multi-Agent Collaboration
• Learning to Ponder: Adaptive Reasoning in Latent Space
• Model Merging Scaling Laws in Large Language Models
• SpecExit: Accelerating Large Reasoning Model via Speculative Exit
• Interactive Program Synthesis for Modeling Collaborative Physical Activities from Narrated Demonstrations
• Rethinking and Benchmarking Large Language Models for Graph Reasoning
• Risk-Sensitive RL for Alleviating Exploration Dilemmas in Large Language Models
• PAME-AI: Patient Messaging Creation and Optimization using Agentic AI
• AdvChain: Adversarial Chain-of-Thought Tuning for Robust Safety Alignment of Large Reasoning Models
• G-reasoner: Foundation Models for Unified Reasoning over Graph-structured Knowledge
• SCI-Verifier: Scientific Verifier with Thinking
• Experience Paper: Adopting Activity Recognition in On-demand Food Delivery Business
• MedMMV: A Controllable Multimodal Multi-Agent Framework for Reliable and Verifiable Clinical Reasoning
• humancompatible.detect: a Python Toolkit for Detecting Bias in AI Models
• Fin-Ally: Pioneering the Development of an Advanced, Commonsense-Embedded Conversational AI for Money Matters
• From Static to Dynamic: Adaptive Monte Carlo Search for Mathematical Process Supervision
• Plan before Solving: Problem-Aware Strategy Routing for Mathematical Reasoning with LLMs
• Towards Safe Reasoning in Large Reasoning Models via Corrective Intervention
• A Systematic Review of Digital Twin-Driven Predictive Maintenance in Industrial Engineering: Taxonomy, Architectural Elements, and Future Research Directions
• ContextPRM: Leveraging Contextual Coherence for multi-domain Test-Time Scaling
• Overcoming Over-Fitting in Constraint Acquisition via Query-Driven Interactive Refinement
• Neuroplasticity-inspired dynamic ANNs for multi-task demand forecasting
• Experience-guided reflective co-evolution of prompts and heuristics for automatic algorithm design
• Training Agents Inside of Scalable World Models
• BPMN Assistant: An LLM-Based Approach to Business Process Modeling
• LTL$_f$ Learning Meets Boolean Set Cover
• "Stop replacing salt with sugar!'': Towards Intuitive Human-Agent Teaching
• Successful Misunderstandings: Learning to Coordinate Without Being Understood
• SysMoBench: Evaluating AI on Formally Modeling Complex Real-World Systems
• MathBode: Frequency-Domain Fingerprints of LLM Mathematical Reasoning
• Coordination Requires Simplification: Thermodynamic Bounds on Multi-Objective Compromise in Natural and Artificial Intelligence
• AI-Enhanced Distributed Channel Access for Collision Avoidance in Future Wi-Fi 8
• Limit Analysis for Symbolic Multi-step Reasoning Tasks with Information Propagation Rules Based on Transformers
• Understanding and Enhancing the Planning Capability of Language Models via Multi-Token Prediction
• AutoEP: LLMs-Driven Automation of Hyperparameter Evolution for Metaheuristic Algorithms
• $p$-less Sampling: A Robust Hyperparameter-Free Approach for LLM Decoding
• Agentic AI Reasoning for Mobile Edge General Intelligence: Fundamentals, Approaches, and Directions
• Training Vision-Language Process Reward Models for Test-Time Scaling in Multimodal Reasoning: Key Insights and Lessons Learned
• GUI-PRA: Process Reward Agent for GUI Tasks
• Socio-Economic Model of AI Agents
• Toward Effective Tool-Integrated Reasoning via Self-Evolved Preference Learning
• Learning How to Use Tools, Not Just When: Pattern-Aware Tool-Integrated Reasoning
• Your Models Have Thought Enough: Training Large Reasoning Models to Stop Overthinking
• From Conversation to Query Execution: Benchmarking User and Tool Interactions for EHR Database Agents
• Democratizing AI scientists using ToolUniverse
• Beyond Embeddings: Interpretable Feature Extraction for Binary Code Similarity
• ViTSP: A Vision Language Models Guided Framework for Large-Scale Traveling Salesman Problems
• GeoBS: Information-Theoretic Quantification of Geographic Bias in AI Models
• Accurate Predictions in Education with Discrete Variational Inference
• Mapping Overlaps in Benchmarks through Perplexity in the Wild
• Dynamic Trust Calibration Using Contextual Bandits
• Model Consistency as a Cheap yet Predictive Proxy for LLM Elo Scores
• DOoM: Difficult Olympiads of Math
• Beyond the Strongest LLM: Multi-Turn Multi-Agent Orchestration vs. Single LLMs on Benchmarks
• Formalization Driven LLM Prompt Jailbreaking via Reinforcement Learning
• A Hierarchical Structure-Enhanced Personalized Recommendation Model for Traditional Chinese Medicine Formulas Based on KG Diffusion Guidance
• Clean First, Align Later: Benchmarking Preference Data Cleaning for Reliable LLM Alignment
• BridgeDrive: Diffusion Bridge Policy for Closed-Loop Trajectory Planning in Autonomous Driving
• PSG-Agent: Personality-Aware Safety Guardrail for LLM-based Agents
• Reasoning Scaffolding: Distilling the Flow of Thought from LLMs
• How LLMs Learn to Reason: A Complex Network Perspective
• Game-Oriented ASR Error Correction via RAG-Enhanced LLM
• From Reasoning to Answer: Empirical, Attention-Based and Mechanistic Insights into Distilled DeepSeek R1 Models
• SafeSearch: Automated Red-Teaming for the Safety of LLM-Based Search Agents
• Measuring Sparse Autoencoder Feature Sensitivity
• MedLA: A Logic-Driven Multi-Agent Framework for Complex Medical Reasoning with Large Language Models
• EAPO: Enhancing Policy Optimization with On-Demand Expert Assistance
• Diagnosing Failure Root Causes in Platform-Orchestrated Agentic Systems: Dataset, Taxonomy, and Benchmark
• GUI-Shepherd: Reliable Process Reward and Verification for Long-Sequence GUI Tasks
• Transparent Visual Reasoning via Object-Centric Agent Collaboration
• From What to Why: A Multi-Agent System for Evidence-based Chemical Reaction Condition Reasoning
• Falcon: A Cross-Modal Evaluation Dataset for Comprehensive Safety Perception
• From Frustration to Fun: An Adaptive Problem-Solving Puzzle Game Powered by Genetic Algorithm
• Mixture-of-Visual-Thoughts: Exploring Context-Adaptive Reasoning Mode Selection for General Visual Reasoning
• Can Large Language Models Develop Gambling Addiction?
• Hilbert: Recursively Building Formal Proofs with Informal Reasoning
• Toward a Theory of Generalizability in LLM Mechanistic Interpretability Research
• JE-IRT: A Geometric Lens on LLM Abilities through Joint Embedding Item Response Theory
• Not only a helper, but also a teacher: Interactive LLM Cascade
• Towards Strategic Persuasion with Language Models
• AI Noether -- Bridging the Gap Between Scientific Laws Derived by AI Systems and Canonical Knowledge via Abductive Inference
• Creative Adversarial Testing (CAT): A Novel Framework for Evaluating Goal-Oriented Agentic AI Systems
• Deceive, Detect, and Disclose: Large Language Models Play Mini-Mafia
• Kimi-Dev: Agentless Training as Skill Prior for SWE-Agents
• Risk Profiling and Modulation for LLMs
• Multiplayer Nash Preference Optimization
• Artificial Phantasia: Evidence for Propositional Reasoning-Based Mental Imagery in Large Language Models
• AttAnchor: Guiding Cross-Modal Token Alignment in VLMs with Attention Anchors
• Exploring LLM-based Frameworks for Fault Diagnosis
• Transferring Vision-Language-Action Models to Industry Applications: Architectures, Performance, and Challenges
• Learning Smooth State-Dependent Traversability from Dense Point Clouds
• ODE-GS: Latent ODEs for Dynamic Scene Extrapolation with 3D Gaussian Splatting
• VisioMath: Benchmarking Figure-based Mathematical Reasoning in LMMs
• Neural-Augmented Kelvinlet for Real-Time Soft Tissue Deformation Modeling
• Meta Pruning via Graph Metanetworks : A Universal Meta Learning Framework for Network Pruning
• Origins of Creativity in Attention-Based Diffusion Models
• RAM-W1K: A Multi-Task Wrist Dataset and Benchmark for Rheumatoid Arthritis
• Warm Starts Accelerate Conditional Diffusion
• Vidar: Embodied Video Diffusion Model for Generalist Manipulation
• The Geometry of Cortical Computation: Manifold Disentanglement and Predictive Dynamics in VCNet
• MoQE: Improve Quantization Model performance via Mixture of Quantization Experts
• 3D-LATTE: Latent Space 3D Editing from Textual Instructions
• Can General-Purpose Omnimodels Compete with Specialists? A Case Study in Medical Image Segmentation
• Graph Alignment via Dual-Pass Spectral Encoding and Latent Space Communication
• WorldForge: Unlocking Emergent 3D/4D Generation in Video Diffusion Model via Training-Free Guidance
• Joint Memory Frequency and Computing Frequency Scaling for Energy-efficient DNN Inference
• HUNT: High-Speed UAV Navigation and Tracking in Unstructured Environments via Instantaneous Relative Frames
• Implicit-ARAP: Efficient Handle-Guided Neural Field Deformation via Local Patch Meshing
• Differential Encoding for Improved Representation Learning over Graphs
• Attentive Dilated Convolution for Automatic Sleep Staging using Force-directed Layout
• Similarity-Dissimilarity Loss for Multi-label Supervised Contrastive Learning
• Chronic Obstructive Pulmonary Disease Prediction Using Deep Convolutional Network
• Freqformer: Frequency-Domain Transformer for 3-D Reconstruction and Quantification of Human Retinal Vasculature
• Towards agile multi-robot systems in the real world: Fast onboard tracking of active blinking markers for relative localization
• Generative Video Semantic Communication via Multimodal Semantic Fusion with Large Model
• GCDance: Genre-Controlled Music-Driven 3D Full Body Dance Generation
• Reconstruct Anything Model: a lightweight foundation model for computational imaging
• AdaRank: Adaptive Rank Pruning for Enhanced Model Merging
• In-2-4D: Inbetweening from Two Single-View Images to 4D Generation
• Sharpness-Aware Minimization with Z-Score Gradient Filtering
• RainPro-8: An Efficient Deep Learning Model to Estimate Rainfall Probabilities Over 8 Hours
• Visual Planning: Let's Think Only with Images
• MINGLE: Mixture of Null-Space Gated Low-Rank Experts for Test-Time Continual Model Merging
• Scaling Diffusion Transformers Efficiently via $\mu$P
• AgentThink: A Unified Framework for Tool-Augmented Chain-of-Thought Reasoning in Vision-Language Models for Autonomous Driving
• Diagnosing and Mitigating Modality Interference in Multimodal Large Language Models
• ART-DECO: Arbitrary Text Guidance for 3D Detailizer Construction
• Vision Language Models are Biased
• Hanfu-Bench: A Multimodal Benchmark on Cross-Temporal Cultural Understanding and Transcreation
• Interaction Field Matching: Overcoming Limitations of Electrostatic Models
• Bridging Semantic Logic Gaps: A Cognition Inspired Multimodal Boundary Preserving Network for Image Manipulation Localization
• BLADE: Block-Sparse Attention Meets Step Distillation for Efficient Video Generation
• G-CUT3R: Guided 3D Reconstruction with Camera and Depth Prior Integration
• Temporal Grounding as a Learning Signal for Referring Video Object Segmentation
• Semantic Discrepancy-aware Detector for Image Forgery Identification
• SpotEdit: Evaluating Visually-Guided Image Editing Methods
• Human-like Content Analysis for Generative AI with Language-Grounded Sparse Encoders
• SemaMIL: Semantic-Aware Multiple Instance Learning with Retrieval-Guided State Space Modeling for Whole Slide Images
• Multimodal Iterative RAG for Knowledge-Intensive Visual Question Answering
• Durian: Dual Reference Image-Guided Portrait Animation with Attribute Transfer
• Physics-Guided Null-Space Diffusion with Sparse Masking for Corrective Sparse-View CT Reconstruction
• BranchGRPO: Stable and Efficient GRPO with Structured Branching in Diffusion Models
• Fracture Detection In X-rays Using Custom Convolutional Neural Network (CNN) And Transfer Learning Models
• DEPFusion: Dual-Domain Enhancement and Priority-Guided Mamba Fusion for UAV Multispectral Object Detection
• Loc$^2$: Interpretable Cross-View Localization via Depth-Lifted Local Feature Matching
• MindVL: Towards Efficient and Effective Training of Multimodal Large Language Models on Ascend NPUs
• Lost in Translation? Vocabulary Alignment for Source-Free Adaptation in Open-Vocabulary Semantic Segmentation
• ViSpec: Accelerating Vision-Language Models with Vision-Aware Speculative Decoding
• Pyramid Token Pruning for High-Resolution Large Vision-Language Models via Region, Token, and Instruction-Guided Importance
• VFRTok: Variable Frame Rates Video Tokenizer with Duration-Proportional Information Assumption
• VisionReasoner: Unified Reasoning-Integrated Visual Perception via Reinforcement Learning
• Vid2World: Crafting Video Diffusion Models to Interactive World Models
• Better Safe Than Sorry? Overreaction Problem of Vision Language Models in Visual Emergency Recognition
• Advancing Marine Research: UWSAM Framework and UIIS10K Dataset for Precise Underwater Instance Segmentation
• OViP: Online Vision-Language Preference Learning for VLM Hallucination
• Pixels Versus Priors: Controlling Knowledge Priors in Vision-Language Models through Visual Counterfacts
• InfoDet: A Dataset for Infographic Element Detection
• T2VUnlearning: A Concept Erasing Method for Text-to-Video Diffusion Models
• Boosting Open Set Recognition Performance through Modulated Representation Learning
• CoT-RVS: Zero-Shot Chain-of-Thought Reasoning Segmentation for Videos
• Mod-Adapter: Tuning-Free and Versatile Multi-concept Personalization via Modulation Adapter
• ReDDiT: Rehashing Noise for Discrete Visual Generation
• GLEAM: Learning Generalizable Exploration Policy for Active Mapping in Complex 3D Indoor Scenes
• Score Replacement with Bounded Deviation for Rare Prompt Generation
• MagicTryOn: Harnessing Diffusion Transformer for Garment-Preserving Video Virtual Try-on
• Any-to-Bokeh: Arbitrary-Subject Video Refocusing with Video Diffusion Model
• Cross-modal RAG: Sub-dimensional Text-to-Image Retrieval-Augmented Generation
• CryoCCD: Conditional Cycle-consistent Diffusion with Biophysical Modeling for Cryo-EM Synthesis
• ProxyThinker: Test-Time Guidance through Small Visual Reasoners
• EgoVIS@CVPR: What Changed and What Could Have Changed? State-Change Counterfactuals for Procedure-Aware Video Representation Learning
• EarthMind: Leveraging Cross-Sensor Data for Advanced Earth Observation Interpretation with a Unified Multimodal LLM
• EgoVIS@CVPR: PAIR-Net: Enhancing Egocentric Speaker Detection via Pretrained Audio-Visual Fusion and Alignment Loss
• METok: Multi-Stage Event-based Token Compression for Efficient Long Video Understanding
• Struct2D: A Perception-Guided Framework for Spatial Reasoning in Large Multimodal Models
• Training-Free Diffusion Framework for Stylized Image Generation with Identity Preservation
• Vision-EKIPL: External Knowledge-Infused Policy Learning for Visual Reasoning
• ReCogDrive: A Reinforced Cognitive Framework for End-to-End Autonomous Driving
• MIRAGE: Multimodal foundation model and benchmark for comprehensive retinal OCT image analysis
• Revisiting Visual Understanding in Multimodal Reasoning through a Lens of Image Perturbation
• DART: Differentiable Dynamic Adaptive Region Tokenizer for Vision Foundation Models
• Decoupled Classifier-Free Guidance for Counterfactual Diffusion Models
• FindingDory: A Benchmark to Evaluate Memory in Embodied Agents
• Do We Need Large VLMs for Spotting Soccer Actions?
• From Drawings to Decisions: A Hybrid Vision-Language Framework for Parsing 2D Engineering Drawings into Structured Manufacturing Knowledge
• Improving Black-Box Generative Attacks via Generator Semantic Consistency
• OmniGen2: Exploration to Advanced Multimodal Generation
• Light of Normals: Unified Feature Representation for Universal Photometric Stereo
• SimpleGVR: A Simple Baseline for Latent-Cascaded Video Super-Resolution
• XTransfer: Modality-Agnostic Few-Shot Model Transfer for Human Sensing at the Edge
• Controllable Reference Guided Diffusion with Local Global Fusion for Real World Remote Sensing Image Super Resolution
• Representation Entanglement for Generation: Training Diffusion Transformers Is Much Easier Than You Think
• FA: Forced Prompt Learning of Vision-Language Models for Out-of-Distribution Detection
• Counterfactual Visual Explanation via Causally-Guided Adversarial Steering
• 3DGAA: Realistic and Robust 3D Gaussian-based Adversarial Attack for Autonomous Driving
• NoiseSDF2NoiseSDF: Learning Clean Neural Fields from Noisy Supervision
• CHROMA: Consistent Harmonization of Multi-View Appearance via Bilateral Grid Prediction
• Disentangling Regional Primitives for Image Generation
• DiffusionGuard: A Robust Defense Against Malicious Diffusion-based Image Editing
• CART: Compositional Auto-Regressive Transformer for Image Generation
• Continuous Speculative Decoding for Autoregressive Image Generation
• Open-Vocabulary Online Semantic Mapping for SLAM
• GenMix: Effective Data Augmentation with Generative Diffusion Model Image Editing
• LoRACLR: Contrastive Adaptation for Customization of Diffusion Models
• Measurement of Medial Elbow Joint Space using Landmark Detection
• Reversed in Time: A Novel Temporal-Emphasized Benchmark for Cross-Modal Video-Text Retrieval
• PERSE: Personalized 3D Generative Avatars from A Single Portrait
• Training-Free Defense Against Adversarial Attacks in Deep Learning MRI Reconstruction
• MIAFEx: An Attention-based Feature Extraction Method for Medical Image Classification
• CGI: Identifying Conditional Generative Models with Example Images
• Med-PU: Point Cloud Upsampling for High-Fidelity 3D Medical Shape Reconstruction
• DeepFRC: An End-to-End Deep Learning Model for Functional Registration and Classification
• Half-order Fine-Tuning for Diffusion Model: A Recursive Likelihood Ratio Optimizer
• 3D Foundation Model for Generalizable Disease Detection in Head Computed Tomography
• PoI: A Filter to Extract Pixel of Interest from Novel View Synthesis for Scene Coordinate Regression
• Bidirectional Uncertainty-Aware Region Learning for Semi-Supervised Medical Image Segmentation
• IM360: Large-scale Indoor Mapping with 360 Cameras
• VPNeXt -- Rethinking Dense Decoding for Plain Vision Transformer
• Spiking Meets Attention: Efficient Remote Sensing Image Super-Resolution with Attention Spiking Neural Networks
• High-Precision Dichotomous Image Segmentation via Depth Integrity-Prior and Fine-Grained Patch Strategy
• Find your Needle: Small Object Image Retrieval via Multi-Object Attention Optimization
• Exploring Reprensentation Invariance in Finetuning
• UniF$^2$ace: A Unified Fine-grained Face Understanding and Generation Model
• Controllable Adversarial Makeup for Privacy via Text-Guided Diffusion
• A Survey on Self-supervised Contrastive Learning for Multimodal Text-Image Analysis
• Interpretable 3D Neural Object Volumes for Robust Conceptual Reasoning
• DPFlow: Adaptive Optical Flow Estimation with a Dual-Pyramid Framework
• Plug-and-Play 1.x-Bit KV Cache Quantization for Video Large Language Models
• Efficient Self-Supervised Adaptation for Medical Image Analysis
• Audio-centric Video Understanding Benchmark without Text Shortcut
• Beyond Synthetic Replays: Turning Diffusion Features into Few-Shot Class-Incremental Learning Knowledge
• SCRAMBLe : Enhancing Multimodal LLM Compositionality with Synthetic Preference Data
• From Specificity to Generality: Revisiting Generalizable Artifacts in Detecting Face Deepfakes
• VideoChat-R1: Enhancing Spatio-Temporal Perception via Reinforcement Fine-Tuning
• Self-alignment of Large Video Language Models with Refined Regularized Preference Optimization
• HSACNet: Hierarchical Scale-Aware Consistency Regularized Semi-Supervised Change Detection
• Model-based Metric 3D Shape and Motion Reconstruction of Wild Bottlenose Dolphins in Drone-Shot Videos
• DreamO: A Unified Framework for Image Customization
• S2S-Net: Addressing the Domain Gap of Heterogeneous Sensor Systems in LiDAR-Based Collective Perception
• ReactDance: Hierarchical Representation for High-Fidelity and Coherent Long-Form Reactive Dance Generation
• Dynamic Uncertainty Learning with Noisy Correspondence for Text-Based Person Search
• QVGen: Pushing the Limit of Quantized Video Generative Models
• ZeroScene: A Zero-Shot Framework for 3D Scene Generation from a Single Image and Controllable Texture Editing
• Efficient Audio-Visual Speech Separation with Discrete Lip Semantics and Multi-Scale Global-Local Attention
• Focusing on What Matters: Object-Agent-centric Tokenization for Vision Language Action models
• DFG-PCN: Point Cloud Completion with Degree-Flexible Point Graph
• StrucADT: Generating Structure-controlled 3D Point Clouds with Adjacency Diffusion Transformer
• Diff-3DCap: Shape Captioning with Diffusion Models
• GBSK: Skeleton Clustering via Granular-ball Computing and Multi-Sampling for Large-Scale Data
• Transparent Visual Reasoning via Object-Centric Agent Collaboration
• Accuracy-Robustness Trade Off via Spiking Neural Network Gradient Sparsity Trail
• ReLumix: Extending Image Relighting to Video via Video Diffusion Models
• FedAgentBench: Towards Automating Real-world Federated Medical Image Analysis with Server-Client LLM Agents
• AISHELL6-whisper: A Chinese Mandarin Audio-visual Whisper Speech Dataset with Speech Recognition Baselines
• Efficient Multi-turn RL for GUI Agents via Decoupled Training and Adaptive Data Curation
• Taught Well Learned Ill: Towards Distillation-conditional Backdoor Attack
• Interpreting deep learning-based stellar mass estimation via causal analysis and mutual information decomposition
• A University of Texas Medical Branch Case Study on Aortic Calcification Detection
• SLA: Beyond Sparsity in Diffusion Transformers via Fine-Tunable Sparse-Linear Attention
• GPS-MTM: Capturing Pattern of Normalcy in GPS-Trajectories with self-supervised learning
• End-to-end Topographic Auditory Models Replicate Signatures of Human Auditory Cortex
• AQUAIR: A High-Resolution Indoor Environmental Quality Dataset for Smart Aquaculture Monitoring
• Clebsch-Gordan Transformer: Fast and Global Equivariant Attention
• Mash, Spread, Slice! Learning to Manipulate Object States via Visual Spatial Progress
• Neural Visibility of Point Sets
• Semantic Editing with Coupled Stochastic Differential Equations
• Non-Invasive Detection of PROState Cancer with Novel Time-Dependent Diffusion MRI and AI-Enhanced Quantitative Radiological Interpretation: PROS-TD-AI
• PROFusion: Robust and Accurate Dense Reconstruction via Camera Pose Regression and Optimization
• Rethinking JEPA: Compute-Efficient Video SSL with Frozen Teachers
• ReCon-GS: Continuum-Preserved Guassian Streaming for Fast and Compact Reconstruction of Dynamic Scenes
• TraitSpaces: Towards Interpretable Visual Creativity for Human-AI Co-Creation
• Wavelet-Assisted Mamba for Satellite-Derived Sea Surface Temperature Super-Resolution
• Hybrid Layer-Wise ANN-SNN With Surrogate Spike Encoding-Decoding Structure
• A Novel Preprocessing Unit for Effective Deep Learning based Classification and Grading of Diabetic Retinopathy
• SAIP: A Plug-and-Play Scale-adaptive Module in Diffusion-based Inverse Problems
• Discovering "Words" in Music: Unsupervised Learning of Compositional Sparse Code for Symbolic Music
• CEDex: Cross-Embodiment Dexterous Grasp Generation at Scale from Human-like Contact Representations
• A TRIANGLE Enables Multimodal Alignment Beyond Cosine Similarity
• VSSFlow: Unifying Video-conditioned Sound and Speech Generation via Joint Learning
• Of-SemWat: High-payload text embedding for semantic watermarking of AI-generated images with arbitrary size
• DRCP: Diffusion on Reinforced Cooperative Perception for Perceiving Beyond Limits
• Light-SQ: Structure-aware Shape Abstraction with Superquadrics for Generated Meshes
• Score-based Membership Inference on Diffusion Models
• Uncertainty-Aware Deep Learning for Wildfire Danger Forecasting
• AIRoA MoMa Dataset: A Large-Scale Hierarchical Dataset for Mobile Manipulation
• CharGen: Fast and Fluent Portrait Modification
• Unsupervised Representation Learning for 3D Mesh Parameterization with Semantic and Visibility Objectives
• MGM-Omni: Scaling Omni LLMs to Personalized Long-Horizon Speech
• LayerD: Decomposing Raster Graphic Designs into Layers
• Vision-and-Language Navigation with Analogical Textual Descriptions in LLMs
• Learning to Infer Unseen Single-/Multi-Attribute-Object Compositions with Graph Networks
• TRIPS: Efficient Vision-and-Language Pre-training with Text-Relevant Image Patch Selection
• Self-Supervised Geometry-Guided Initialization for Robust Monocular Visual Odometry
• fVDB: A Deep-Learning Framework for Sparse, Large-Scale, and High-Performance Spatial Intelligence
• Leveraging BEV Paradigm for Ground-to-Aerial Image Synthesis
• VT-FSL: Bridging Vision and Text with LLMs for Few-Shot Learning
• Fast Real-Time Pipeline for Robust Arm Gesture Recognition
• A Scalable Distributed Framework for Multimodal GigaVoxel Image Registration
• GEM: 3D Gaussian Splatting for Efficient and Accurate Cryo-EM Reconstruction
• BRIDGE - Building Reinforcement-Learning Depth-to-Image Data Generation Engine for Monocular Depth Estimation
• UniLat3D: Geometry-Appearance Unified Latents for Single-Stage 3D Generation
• MANI-Pure: Magnitude-Adaptive Noise Injection for Adversarial Purification
• Triangle Splatting+: Differentiable Rendering with Opaque Triangles
• Score Distillation of Flow Matching Models
• TemMed-Bench: Evaluating Temporal Medical Image Reasoning in Vision-Language Models
• Fast Feature Field ($\text{F}^3$): A Predictive Representation of Events
• VideoAnchor: Reinforcing Subspace-Structured Visual Cues for Coherent Visual-Spatial Reasoning
• GSM8K-V: Can Vision Language Models Solve Grade School Math Word Problems in Visual Contexts
• Rolling Forcing: Autoregressive Long Video Diffusion in Real Time
• Aligning Visual Foundation Encoders to Tokenizers for Diffusion Models
• YOLO26: Key Architectural Enhancements and Performance Benchmarking for Real-Time Object Detection
• Personalized Vision via Visual In-Context Learning
• Mitigating Hallucination in Multimodal LLMs with Layer Contrastive Decoding
• GHOST: Hallucination-Inducing Image Generation for Multimodal LLMs
• DC-Gen: Post-Training Diffusion Acceleration with Deeply Compressed Latent Space
• DC-VideoGen: Efficient Video Generation with Deep Compression Video Autoencoder
• PAD3R: Pose-Aware Dynamic 3D Reconstruction from Casual Videos
• PixelCraft: A Multi-Agent System for High-Fidelity Visual Reasoning on Structured Images
• FlashI2V: Fourier-Guided Latent Shifting Prevents Conditional Image Leakage in Image-to-Video Generation
• Visual Jigsaw Post-Training Improves MLLMs
• VGGT-X: When VGGT Meets Dense Novel View Synthesis
• Q2E: Query-to-Event Decomposition for Zero-Shot Multilingual Text-to-Video Retrieval
• YOLO-based Bearing Fault Diagnosis With Continuous Wavelet Transform
• VIRTUS-FPP: Virtual Sensor Modeling for Fringe Projection Profilometry in NVIDIA Isaac Sim
• ReSeFlow: Rectifying SE(3)-Equivariant Policy Learning Flows
• Explainable Deep Learning for Cataract Detection in Retinal Images: A Dual-Eye and Knowledge Distillation Approach
• Localizing Adversarial Attacks To Produces More Imperceptible Noise
• Achieving Fair Skin Lesion Detection through Skin Tone Normalization and Channel Pruning
• Responsible Diffusion: A Comprehensive Survey on Safety, Ethics, and Trust in Diffusion Models
• Consistency Models as Plug-and-Play Priors for Inverse Problems
• Mixture-of-Visual-Thoughts: Exploring Context-Adaptive Reasoning Mode Selection for General Visual Reasoning
• Self-driving cars: Are we there yet?
• Introducing Multimodal Paradigm for Learning Sleep Staging PSG via General-Purpose Model
• MonoCon: A general framework for learning ultra-compact high-fidelity representations using monotonicity constraints
• LLMs Behind the Scenes: Enabling Narrative Scene Illustration
• Robot Learning from Any Images
• ADAM: A Diverse Archive of Mankind for Evaluating and Enhancing LLMs in Biographical Reasoning
• UniPrototype: Humn-Robot Skill Learning with Uniform Prototypes
• AttAnchor: Guiding Cross-Modal Token Alignment in VLMs with Attention Anchors
• Leave No Observation Behind: Real-time Correction for VLA Action Chunks
• Training Vision-Language Process Reward Models for Test-Time Scaling in Multimodal Reasoning: Key Insights and Lessons Learned
• Robust Fine-Tuning from Non-Robust Pretrained Models: Mitigating Suboptimal Transfer With Adversarial Scheduling
• Targeted perturbations reveal brain-like local coding axes in robustified, but not standard, ANN-based brain models
• DiffTex: Differentiable Texturing for Architectural Proxy Models
• Graph Your Own Prompt
• CCD: Mitigating Hallucinations in Radiology MLLMs via Clinical Contrastive Decoding
• S$^3$F-Net: A Multi-Modal Approach to Medical Image Classification via Spatial-Spectral Summarizer Fusion Network
• Temporal Generalization: A Reality Check
• RAVEN: Resilient Aerial Navigation via Open-Set Semantic Memory and Behavior Adaptation
• Automated design of compound lenses with discrete-continuous optimization
• BridgeDrive: Diffusion Bridge Policy for Closed-Loop Trajectory Planning in Autonomous Driving
• StolenLoRA: Exploring LoRA Extraction Attacks via Synthetic Data
• Perceive, Reflect and Understand Long Video: Progressive Multi-Granular Clue Exploration with Interactive Agents
• Evaluating Temperature Scaling Calibration Effectiveness for CNNs under Varying Noise Levels in Brain Tumour Detection
• Social 3D Scene Graphs: Modeling Human Actions and Relations for Interactive Service Robots
• Event-based Facial Keypoint Alignment via Cross-Modal Fusion Attention and Self-Supervised Multi-Event Representation Learning
• On-the-Fly Data Augmentation for Brain Tumor Segmentation
• Wan-Alpha: High-Quality Text-to-Video Generation with Alpha Channel
• SDPose: Exploiting Diffusion Priors for Out-of-Domain and Robust Pose Estimation
• PanoWorld-X: Generating Explorable Panoramic Worlds via Sphere-Aware Video Diffusion
• LVT: Large-Scale Scene Reconstruction via Local View Transformers
• CLASP: Adaptive Spectral Clustering for Unsupervised Per-Image Segmentation
• GeoVLM-R1: Reinforcement Fine-Tuning for Improved Remote Sensing Reasoning
• STAGE: Stable and Generalizable GRPO for Autoregressive Image Generation
• TokenSwap: Backdoor Attack on the Compositional Understanding of Large Vision-Language Models
• SCOPE: Semantic Conditioning for Sim2Real Category-Level Object Pose Estimation in Robotics
• BFSM: 3D Bidirectional Face-Skull Morphable Model
• Comprehensive Benchmarking of YOLOv11 Architectures for Scalable and Granular Peripheral Blood Cell Detection
• Biomechanical-phase based Temporal Segmentation in Sports Videos: a Demonstration on Javelin-Throw
• FreeRet: MLLMs as Training-Free Retrievers
• Can you SPLICE it together? A Human Curated Benchmark for Probing Visual Reasoning in VLMs
• RIFLE: Removal of Image Flicker-Banding via Latent Diffusion Enhancement
• Learning Object-Centric Representations Based on Slots in Real World Scenarios
• VNODE: A Piecewise Continuous Volterra Neural Network
• Classifier-Centric Adaptive Framework for Open-Vocabulary Camouflaged Object Segmentation
• Traumatic Brain Injury Segmentation using an Ensemble of Encoder-decoder Models
• SANA-Video: Efficient Video Generation with Block Linear Diffusion Transformer
• Enhancing Physical Plausibility in Video Generation by Reasoning the Implausibility
• IWR-Bench: Can LVLMs reconstruct interactive webpage from a user interaction video?
• Evaluation of Polarimetric Fusion for Semantic Segmentation in Aquatic Environments
• Toward a Vision-Language Foundation Model for Medical Data: Multimodal Dataset and Benchmarks for Vietnamese PET/CT Report Generation
• Collaborating Vision, Depth, and Thermal Signals for Multi-Modal Tracking: Dataset and Algorithm
• ExGS: Extreme 3D Gaussian Compression with Diffusion Priors
• VTPerception-R1: Enhancing Multimodal Reasoning via Explicit Visual and Textual Perceptual Grounding
• SkyLink: Unifying Street-Satellite Geo-Localization via UAV-Mediated 3D Scene Alignment
• LOVE-R1: Advancing Long Video Understanding with an Adaptive Zoom-in Mechanism via Multi-Step Reasoning
• Vision Function Layer in Multimodal LLMs
• Causal-Adapter: Taming Text-to-Image Diffusion for Faithful Counterfactual Generation
• TACO-Net: Topological Signatures Triumph in 3D Object Classification
• UP2You: Fast Reconstruction of Yourself from Unconstrained Photo Collections
• Training-Free Token Pruning via Zeroth-Order Gradient Estimation in Vision-Language Models
• PHASE-Net: Physics-Grounded Harmonic Attention System for Efficient Remote Photoplethysmography Measurement
• ELPG-DTFS: Prior-Guided Adaptive Time-Frequency Graph Neural Network for EEG Depression Diagnosis
• Vision At Night: Exploring Biologically Inspired Preprocessing For Improved Robustness Via Color And Contrast Transformations
• StreamForest: Efficient Online Video Understanding with Persistent Event Memory
• Environment-Aware Satellite Image Generation with Diffusion Models
• ThermalGen: Style-Disentangled Flow-Based Generative Models for RGB-to-Thermal Image Translation
• Vehicle Classification under Extreme Imbalance: A Comparative Study of Ensemble Learning and CNNs
• MMRQA: Signal-Enhanced Multimodal Large Language Models for MRI Quality Assessment
• VAGUEGAN: Stealthy Poisoning and Backdoor Attacks on Image Generative Pipelines
• DWGS: Enhancing Sparse-View Gaussian Splatting with Hybrid-Loss Depth Estimation and Bidirectional Warping
• DAM: Dual Active Learning with Multimodal Foundation Model for Source-Free Domain Adaptation
• Accurate Cobb Angle Estimation via SVD-Based Curve Detection and Vertebral Wedging Quantification
• Attention Surgery: An Efficient Recipe to Linearize Your Video Diffusion Transformer
• OpenGPT-4o-Image: A Comprehensive Dataset for Advanced Image Generation and Editing
• Learning Goal-Oriented Language-Guided Navigation with Self-Improving Demonstrations at Scale
• Segmentor-Guided Counterfactual Fine-Tuning for Image Synthesis
• Scalable GANs with Transformers
• Combining Discrepancy-Confusion Uncertainty and Calibration Diversity for Active Fine-Grained Image Classification
• Tumor Synthesis conditioned on Radiomics
• Simulating Post-Neoadjuvant Chemotherapy Breast Cancer MRI via Diffusion Model with Prompt Tuning
• Talk in Pieces, See in Whole: Disentangling and Hierarchical Aggregating Representations for Language-based Object Detection
• An Efficient 3D Latent Diffusion Model for T1-contrast Enhanced MRI Generation
• UniVid: The Open-Source Unified Video Model
• BALR-SAM: Boundary-Aware Low-Rank Adaptation of SAM for Resource-Efficient Medical Image Segmentation
• Forge4D: Feed-Forward 4D Human Reconstruction and Interpolation from Uncalibrated Sparse-view Videos
• Scalable Audio-Visual Masked Autoencoders for Efficient Affective Video Facial Analysis
• EVLF-FM: Explainable Vision Language Foundation Model for Medicine
• FreeAction: Training-Free Techniques for Enhanced Fidelity of Trajectory-to-Video Generation
• Latent Visual Reasoning
• When MLLMs Meet Compression Distortion: A Coding Paradigm Tailored to MLLMs
• S$^2$NN: Sub-bit Spiking Neural Networks
• Cycle Diffusion Model for Counterfactual Image Generation
• Skeleton-based Robust Registration Framework for Corrupted 3D Point Clouds
• Robust Partial 3D Point Cloud Registration via Confidence Estimation under Global Context
• ASIA: Adaptive 3D Segmentation using Few Image Annotations
• SVGThinker: Instruction-Aligned and Reasoning-Driven Text-to-SVG Generation
• FrameThinker: Learning to Think with Long Videos via Multi-Turn Frame Spotlighting
• OMeGa: Joint Optimization of Explicit Meshes and Gaussian Splats for Robust Scene-Level Surface Reconstruction
• Towards Foundation Models for Cryo-ET Subtomogram Analysis
• Similarity-Aware Selective State-Space Modeling for Semantic Correspondence
• TP-MVCC: Tri-plane Multi-view Fusion Model for Silkie Chicken Counting
• Hyperspherical Latents Improve Continuous-Token Autoregressive Generation
• Dynamic Orchestration of Multi-Agent System for Real-World Multi-Image Agricultural VQA
• NeRV-Diffusion: Diffuse Implicit Neural Representations for Video Synthesis
• An Enhanced Pyramid Feature Network Based on Long-Range Dependencies for Multi-Organ Medical Image Segmentation
• DRIFT: Divergent Response in Filtered Transformations for Robust Adversarial Defense
• UI-UG: A Unified MLLM for UI Understanding and Generation
• Uni-X: Mitigating Modality Conflict with a Two-End-Separated Architecture for Unified Multimodal Models
• Real-Aware Residual Model Merging for Deepfake Detection
• From Satellite to Street: A Hybrid Framework Integrating Stable Diffusion and PanoGAN for Consistent Cross-View Synthesis
• DINOReg: Strong Point Cloud Registration with Vision Foundation Model
• Mask Clustering-based Annotation Engine for Large-Scale Submeter Land Cover Mapping
• REALIGN: Regularized Procedure Alignment with Matching Video Embeddings via Partial Gromov-Wasserstein Optimal Transport
• Vid-LLM: A Compact Video-based 3D Multimodal LLM with Reconstruction-Reasoning Synergy
• PCICF: A Pedestrian Crossing Identification and Classification Framework
• RapidMV: Leveraging Spatio-Angular Representations for Efficient and Consistent Text-to-Multi-View Synthesis
• CLQ: Cross-Layer Guided Orthogonal-based Quantization for Diffusion Transformers
• A Data-Centric Perspective on the Influence of Image Data Quality in Machine Learning Models
• Proxy-GS: Efficient 3D Gaussian Splatting via Proxy Mesh
• Rethinking Unsupervised Cross-modal Flow Estimation: Learning from Decoupled Optimization and Consistency Constraint
• UI2V-Bench: An Understanding-based Image-to-video Generation Benchmark
• NeoWorld: Neural Simulation of Explorable Virtual Worlds via Progressive 3D Unfolding
• Beyond Isolated Facts: Synthesizing Narrative and Grounded Supervision for VideoQA
• Generalist Multi-Class Anomaly Detection via Distillation to Two Heterogeneous Student Networks
• LaMoGen: Laban Movement-Guided Diffusion for Text-to-Motion Generation
• Euclid's Gift: Enhancing Spatial Perception and Reasoning in Vision-Language Models via Geometric Surrogate Tasks
• Performance-Efficiency Trade-off for Fashion Image Retrieval
• Mitigating Visual Hallucinations via Semantic Curriculum Preference Optimization in MLLMs
• Robust Multimodal Semantic Segmentation with Balanced Modality Contributions
• Instruction Guided Multi Object Image Editing with Quantity and Layout Consistency
• CMT: Mid-Training for Efficient Learning of Consistency, Mean Flow, and Flow Map Models
• CORE-3D: Context-aware Open-vocabulary Retrieval by Embeddings in 3D
• Diffusion Bridge or Flow Matching? A Unifying Framework and Comparative Analysis
• Foggy Crowd Counting: Combining Physical Priors and KAN-Graph
• NeMo: Needle in a Montage for Video-Language Understanding
• Assessing Visual Privacy Risks in Multimodal AI: A Novel Taxonomy-Grounded Evaluation of Vision-Language Models
• Uni4D-LLM: A Unified SpatioTemporal-Aware VLM for 4D Understanding and Generation
• 2nd Place Report of MOSEv2 Challenge 2025: Concept Guided Video Object Segmentation via SeC
• Towards Fine-Grained Text-to-3D Quality Assessment: A Benchmark and A Two-Stage Rank-Learning Metric
• CE-FAM: Concept-Based Explanation via Fusion of Activation Maps
• FairViT-GAN: A Hybrid Vision Transformer with Adversarial Debiasing for Fair and Explainable Facial Beauty Prediction
• Sim-DETR: Unlock DETR for Temporal Sentence Grounding
• Not All Tokens are Guided Equal: Improving Guidance in Visual Autoregressive Models
• PCRI: Measuring Context Robustness in Multimodal Models for Enterprise Applications
• Learning Adaptive Pseudo-Label Selection for Semi-Supervised 3D Object Detection
• Tunable-Generalization Diffusion Powered by Self-Supervised Contextual Sub-Data for Low-Dose CT Reconstruction
• AssemblyHands-X: Modeling 3D Hand-Body Coordination for Understanding Bimanual Human Activities
• LifeCLEF Plant Identification Task 2015
• Preserving Cross-Modal Stability for Visual Unlearning in Multimodal Scenarios
• Q-FSRU: Quantum-Augmented Frequency-Spectral For Medical Visual Question Answering
• LifeCLEF Plant Identification Task 2014
• EWC-Guided Diffusion Replay for Exemplar-Free Continual Learning in Medical Imaging
• Adversarial Versus Federated: An Adversarial Learning based Multi-Modality Cross-Domain Federated Medical Segmentation
• EditScore: Unlocking Online RL for Image Editing via High-Fidelity Reward Modeling
• MoReact: Generating Reactive Motion from Textual Descriptions
• Revisit the Imbalance Optimization in Multi-task Learning: An Experimental Analysis
• Bridging the Task Gap: Multi-Task Adversarial Transferability in CLIP and Its Derivatives
• Token Painter: Training-Free Text-Guided Image Inpainting via Mask Autoregressive Models
• DriveE2E: Closed-Loop Benchmark for End-to-End Autonomous Driving through Real-to-Simulation
• Learning Encoding-Decoding Direction Pairs to Unveil Concepts of Influence in Deep Vision Networks
• SAR-KnowLIP: Towards Multimodal Foundation Models for Remote Sensing
• AutoPrune: Each Complexity Deserves a Pruning Policy
• CrashSplat: 2D to 3D Vehicle Damage Segmentation in Gaussian Splatting
• HunyuanImage 3.0 Technical Report
• ColLab: A Collaborative Spatial Progressive Data Engine for Referring Expression Comprehension and Generation
• Reinforcement Learning with Inverse Rewards for World Model Post-training
• A Novel Hybrid Deep Learning and Chaotic Dynamics Approach for Thyroid Cancer Classification
• VFSI: Validity First Spatial Intelligence for Constraint-Guided Traffic Diffusion
• Towards Redundancy Reduction in Diffusion Models for Efficient Video Super-Resolution
• RPG360: Robust 360 Depth Estimation with Perspective Foundation Models and Graph Optimization
• Advancing Multi-agent Traffic Simulation via R1-Style Reinforcement Fine-Tuning
• TREAT-Net: Tabular-Referenced Echocardiography Analysis for Acute Coronary Syndrome Treatment Prediction
• Gaze Estimation for Human-Robot Interaction: Analysis Using the NICO Platform
• SIE3D: Single-image Expressive 3D Avatar generation via Semantic Embedding and Perceptual Expression Loss
• FrameMind: Frame-Interleaved Chain-of-Thought for Video Reasoning via Reinforcement Learning
• Generalized Category Discovery in Hyperspectral Images via Prototype Subspace Modeling
• Hazy Pedestrian Trajectory Prediction via Physical Priors and Graph-Mamba
• $\mathbf{R}^3$: Reconstruction, Raw, and Rain: Deraining Directly in the Bayer Domain
• Joint Superpixel and Self-Representation Learning for Scalable Hyperspectral Image Clustering
• A Second-Order Perspective on Pruning at Initialization and Knowledge Transfer
• Uncovering Grounding IDs: How External Cues Shape Multi-Modal Binding
• Autoregressive Video Generation beyond Next Frames Prediction
• Unified Multi-Modal Interactive & Reactive 3D Motion Generation via Rectified Flow
• SVAC: Scaling Is All You Need For Referring Video Object Segmentation
• GANji: A Framework for Introductory AI Image Generation
• Generalist Scanner Meets Specialist Locator: A Synergistic Coarse-to-Fine Framework for Robust GUI Grounding
• EYE-DEX: Eye Disease Detection and EXplanation System
• Analysis of Bias in Deep Learning Facial Beauty Regressors
• Asymmetric VAE for One-Step Video Super-Resolution Acceleration
• Accelerating Cerebral Diagnostics with BrainFusion: A Comprehensive MRI Tumor Framework
• LatXGen: Towards Radiation-Free and Accurate Quantitative Analysis of Sagittal Spinal Alignment Via Cross-Modal Radiographic View Synthesis
• High-Order Progressive Trajectory Matching for Medical Image Dataset Distillation
• Evaluating point-light biological motion in multimodal large language models
• Imaging-Based Mortality Prediction in Patients with Systemic Sclerosis
• Calibrated and Resource-Aware Super-Resolution for Reliable Driver Behavior Analysis
• OVSeg3R: Learn Open-vocabulary Instance Segmentation from 2D via 3D Reconstruction
• From Fields to Splats: A Cross-Domain Survey of Real-Time Neural Scene Representations
• Pancreas Part Segmentation under Federated Learning Paradigm
• Towards Interpretable Visual Decoding with Attention to Brain Representations
• RobuQ: Pushing DiTs to W1.58A2 via Robust Activation Quantization
• VividFace: High-Quality and Efficient One-Step Diffusion For Video Face Enhancement
• Multi-Level Heterogeneous Knowledge Transfer Network on Forward Scattering Center Model for Limited Samples SAR ATR
• VAMamba: An Efficient Visual Adaptive Mamba for Image Restoration
• Deep Taxonomic Networks for Unsupervised Hierarchical Prototype Discovery
• MAN: Latent Diffusion Enhanced Multistage Anti-Noise Network for Efficient and High-Quality Low-Dose CT Image Denoising
• VMDiff: Visual Mixing Diffusion for Limitless Cross-Object Synthesis
• FlowLUT: Efficient Image Enhancement via Differentiable LUTs and Iterative Flow Matching
• InteractMove: Text-Controlled Human-Object Interaction Generation in 3D Scenes with Movable Objects
• BioVessel-Net and RetinaMix: Unsupervised Retinal Vessel Segmentation from OCTA Images
• DiffInk: Glyph- and Style-Aware Latent Diffusion Transformer for Text to Online Handwriting Generation
• RIV: Recursive Introspection Mask Diffusion Vision Language Model
• Efficient Domain-Adaptive Multi-Task Dense Prediction with Vision Foundation Models
• MotionVerse: A Unified Multimodal Framework for Motion Comprehension, Generation and Editing
• LightFair: Towards an Efficient Alternative for Fair T2I Diffusion via Debiasing Pre-trained Text Encoders
• EfficientMIL: Efficient Linear-Complexity MIL Method for WSI Classification
• From Static to Dynamic: a Survey of Topology-Aware Perception in Autonomous Driving
• Griffin: Generative Reference and Layout Guided Image Composition
• Sparse-Up: Learnable Sparse Upsampling for 3D Generation with High-Fidelity Textures
• Color-Pair Guided Robust Zero-Shot 6D Pose Estimation and Tracking of Cluttered Objects on Edge Devices
• ReWatch-R1: Boosting Complex Video Reasoning in Large Vision-Language Models through Agentic Data Synthesis
• LLaVA-OneVision-1.5: Fully Open Framework for Democratized Multimodal Training
• HIVTP: A Training-Free Method to Improve VLMs Efficiency via Hierarchical Visual Token Pruning Using Middle-Layer-Based Importance Score
• Token Merging via Spatiotemporal Information Mining for Surgical Video Understanding
• RCI: A Score for Evaluating Global and Local Reasoning in Multimodal Benchmarks
• MSD-KMamba: Bidirectional Spatial-Aware Multi-Modal 3D Brain Segmentation via Multi-scale Self-Distilled Fusion Strategy
• QuantSparse: Comprehensively Compressing Video Diffusion Transformer with Model Quantization and Attention Sparsification
• HomeSafeBench: A Benchmark for Embodied Vision-Language Models in Free-Exploration Home Safety Inspection
• Confidence Aware SSD Ensemble with Weighted Boxes Fusion for Weapon Detection
• INSTINCT: Instance-Level Interaction Architecture for Query-Based Collaborative Perception
• CrimEdit: Controllable Editing for Counterfactual Object Removal, Insertion, and Movement
• PD-Diag-Net: Clinical-Priors guided Network on Brain MRI for Auxiliary Diagnosis of Parkinson's Disease
• DiffPCN: Latent Diffusion Model Based on Multi-view Depth Images for Point Cloud Completion
• Video Panels for Long Video Understanding
• M3DLayout: A Multi-Source Dataset of 3D Indoor Layouts and Structured Descriptions for 3D Generation
• LUQ: Layerwise Ultra-Low Bit Quantization for Multimodal Large Language Models
• FastViDAR: Real-Time Omnidirectional Depth Estimation via Alternative Hierarchical Attention
• HieraTok: Multi-Scale Visual Tokenizer Improves Image Reconstruction and Generation
• GRS-SLAM3R: Real-Time Dense SLAM with Gated Recurrent State
• ResAD++: Towards Class Agnostic Anomaly Detection via Residual Feature Learning
• Poivre: Self-Refining Visual Pointing with Reinforcement Learning
• PVTAdpNet: Polyp Segmentation using Pyramid vision transformer with a novel Adapter block
• UniAlignment: Semantic Alignment for Unified Image Generation, Understanding, Manipulation and Perception
• GenView++: Unifying Adaptive View Generation and Quality-Driven Supervision for Contrastive Representation Learning
• A Modality-Tailored Graph Modeling Framework for Urban Region Representation via Contrastive Learning
• Texture Vector-Quantization and Reconstruction Aware Prediction for Generative Super-Resolution
• GroupCoOp: Group-robust Fine-tuning via Group Prompt Learning
• From Unstable to Playable: Stabilizing Angry Birds Levels via Object Segmentation
• Controllable Generation of Large-Scale 3D Urban Layouts with Semantic and Structural Guidance
• A Multi-Camera Vision-Based Approach for Fine-Grained Assembly Quality Control
• Real-World Transferable Adversarial Attack on Face-Recognition Systems
• UltraUNet: Real-Time Ultrasound Tongue Segmentation for Diverse Linguistic and Imaging Conditions
• Patch Rebirth: Toward Fast and Transferable Model Inversion of Vision Transformers
• Self-Consistency as a Free Lunch: Reducing Hallucinations in Vision-Language Models via Self-Reflection
• TATTOO: Training-free AesTheTic-aware Outfit recOmmendation
• Increasing the Diversity in RGB-to-Thermal Image Translation for Automotive Applications
• LiDAR-based Human Activity Recognition through Laplacian Spectral Analysis
• OracleGS: Grounding Generative Priors for Sparse-View Gaussian Splatting
• Learning Regional Monsoon Patterns with a Multimodal Attention U-Net
• SynDoc: A Hybrid Discriminative-Generative Framework for Enhancing Synthetic Domain-Adaptive Document Key Information Extraction
• Vid-Freeze: Protecting Images from Malicious Image-to-Video Generation via Temporal Freezing
• Seeing Through the Blur: Unlocking Defocus Maps for Deepfake Detection
• Seeing the Unseen in Low-light Spike Streams
• Balanced Diffusion-Guided Fusion for Multimodal Remote Sensing Classification
• Seeing Symbols, Missing Cultures: Probing Vision-Language Models' Reasoning on Fire Imagery and Cultural Meaning
• C3-OWD: A Curriculum Cross-modal Contrastive Learning Framework for Open-World Detection
• Spatial-Spectral Binarized Neural Network for Panchromatic and Multi-spectral Images Fusion
• Decoupling Reasoning and Perception: An LLM-LMM Framework for Faithful Visual Reasoning
• DDP: Dual-Decoupled Prompting for Multi-Label Class-Incremental Learning
• LRPO: Enhancing Blind Face Restoration through Online Reinforcement Learning
• DentVLM: A Multimodal Vision-Language Model for Comprehensive Dental Diagnosis and Enhanced Clinical Practice
• Dynamic-TreeRPO: Breaking the Independent Trajectory Bottleneck with Structured Sampling
• Test-time Uncertainty Estimation for Medical Image Registration via Transformation Equivariance
• GRAPE: Let GPRO Supervise Query Rewriting by Ranking for Retrieval
• CasPoinTr: Point Cloud Completion with Cascaded Networks and Knowledge Distillation
• UniPose: Unified Cross-modality Pose Prior Propagation towards RGB-D data for Weakly Supervised 3D Human Pose Estimation
• Generative Modeling of Shape-Dependent Self-Contact Human Poses
• WorldSplat: Gaussian-Centric Feed-Forward 4D Scene Generation for Autonomous Driving
• Enhanced Fracture Diagnosis Based on Critical Regional and Scale Aware in YOLO
• FracDetNet: Advanced Fracture Detection via Dual-Focus Attention and Multi-scale Calibration in Medical X-ray Imaging
• SPIKE-RL: Video-LLMs meet Bayesian Surprise
• FM-SIREN & FM-FINER: Nyquist-Informed Frequency Multiplier for Implicit Neural Representation with Periodic Activation
• FoR-SALE: Frame of Reference-guided Spatial Adjustment in LLM-based Diffusion Editing
• 3DPCNet: Pose Canonicalization for Robust Viewpoint-Invariant 3D Kinematic Analysis from Monocular RGB cameras
• No Concept Left Behind: Test-Time Optimization for Compositional Text-to-Image Generation
• Robust Multi-Modal Face Anti-Spoofing with Domain Adaptation: Tackling Missing Modalities, Noisy Pseudo-Labels, and Model Degradation
• RestoRect: Degraded Image Restoration via Latent Rectified Flow & Feature Distillation
• Orientation-anchored Hyper-Gaussian for 4D Reconstruction from Casual Videos
• Multi-modal Data Spectrum: Multi-modal Datasets are Multi-dimensional
• Enhancing Polyp Segmentation via Encoder Attention and Dynamic Kernel Update
• Mask What Matters: Controllable Text-Guided Masking for Self-Supervised Medical Image Analysis
• FMC-DETR: Frequency-Decoupled Multi-Domain Coordination for Aerial-View Object Detection
• Follow-Your-Preference: Towards Preference-Aligned Image Inpainting
• Streamline pathology foundation model by cross-magnification distillation
• CoPatch: Zero-Shot Referring Image Segmentation by Leveraging Untapped Spatial Knowledge in CLIP
• Deep Learning for Oral Health: Benchmarking ViT, DeiT, BEiT, ConvNeXt, and Swin Transformer
• HTMA-Net: Towards Multiplication-Avoiding Neural Networks via Hadamard Transform and In-Memory Computing
• Towards Comprehensive Interactive Change Understanding in Remote Sensing: A Large-scale Dataset and Dual-granularity Enhanced VLM
• Stochastic Interpolants via Conditional Dependent Coupling
• Benchmarking DINOv3 for Multi-Task Stroke Analysis on Non-Contrast CT
• Earth-Agent: Unlocking the Full Landscape of Earth Observation with Agents
• WeatherCycle: Unpaired Multi-Weather Restoration via Color Space Decoupled Cycle Learning
• Sparse2Dense: A Keypoint-driven Generative Framework for Human Video Compression and Vertex Prediction
• TRAX: TRacking Axles for Accurate Axle Count Estimation
• Confidence-Calibrating Regularization for Robust Brain MRI Segmentation Under Domain Shift
• Unsupervised Online 3D Instance Segmentation with Synthetic Sequences and Dynamic Loss
• Pathological Truth Bias in Vision-Language Models
• Scale and Rotation Estimation of Similarity-Transformed Images via Cross-Correlation Maximization Based on Auxiliary Function Method
• Robust Object Detection for Autonomous Driving via Curriculum-Guided Group Relative Policy Optimization
• Graph-Theoretic Consistency for Robust and Topology-Aware Semi-Supervised Histopathology Segmentation
• A review of Recent Techniques for Person Re-Identification
• Sequential Token Merging: Revisiting Hidden States
• Deep Learning Empowered Super-Resolution: A Comprehensive Survey and Future Prospects
• Learning Hyperspectral Images with Curated Text Prompts for Efficient Multimodal Alignment
• Global Prompt Refinement with Non-Interfering Attention Masking for One-Shot Federated Learning
• GZSL-MoE: Apprentissage G{\'e}n{\'e}ralis{\'e} Z{\'e}ro-Shot bas{\'e} sur le M{\'e}lange d'Experts pour la Segmentation S{\'e}mantique de Nuages de Points 3DAppliqu{\'e} {\`a} un Jeu de Donn{\'e}es d'Environnement de Collaboration Humain-Robot
• IBiT: Utilizing Inductive Biases to Create a More Data Efficient Attention Mechanism
• LayoutAgent: A Vision-Language Agent Guided Compositional Diffusion for Spatial Layout Planning
• CompareBench: A Benchmark for Visual Comparison Reasoning in Vision-Language Models
• MILR: Improving Multimodal Image Generation via Test-Time Latent Reasoning
• UESA-Net: U-Shaped Embedded Multidirectional Shrinkage Attention Network for Ultrasound Nodule Segmentation
• PartCo: Part-Level Correspondence Priors Enhance Category Discovery
• DEFT: Decompositional Efficient Fine-Tuning for Text-to-Image Models
• VideoScore2: Think before You Score in Generative Video Evaluation
• TRUST: Test-Time Refinement using Uncertainty-Guided SSM Traverses
• MMPB: It's Time for Multi-Modal Personalization
• Seeing Isn't Believing: Context-Aware Adversarial Patch Synthesis via Conditional GAN
• Learning Temporal Saliency for Time Series Forecasting with Cross-Scale Attention
• Multimodal Slice Interaction Network Enhanced by Transfer Learning for Precise Segmentation of Internal Gross Tumor Volume in Lung Cancer PET/CT Imaging
• ControlEvents: Controllable Synthesis of Event Camera Datawith Foundational Prior from Image Diffusion Models
• Learning KAN-based Implicit Neural Representations for Deformable Image Registration
• Convolutional Set Transformer
• TY-RIST: Tactical YOLO Tricks for Real-time Infrared Small Target Detection
• Learning Unified Representation of 3D Gaussian Splatting
• Soft-Di[M]O: Improving One-Step Discrete Image Generation with Soft Embeddings
• FishAI 2.0: Marine Fish Image Classification with Multi-modal Few-shot Learning
• Brain Tumor Classification from MRI Scans via Transfer Learning and Enhanced Feature Representation
• Hemorica: A Comprehensive CT Scan Dataset for Automated Brain Hemorrhage Classification, Segmentation, and Detection
• ARSS: Taming Decoder-only Autoregressive Visual Generation for View Synthesis From Single View
• Disentangling Static and Dynamic Information for Reducing Static Bias in Action Recognition
• Desensitizing for Improving Corruption Robustness in Point Cloud Classification through Adversarial Training
• Geometry-Aware Losses for Structure-Preserving Text-to-Sign Language Generation
• Planning with Unified Multimodal Models
• Copyright Infringement Detection in Text-to-Image Diffusion Models via Differential Privacy
• Perceptual Influence: Improving the Perceptual Loss Design for Low-Dose CT Enhancement
• Sensor-Adaptive Flood Mapping with Pre-trained Multi-Modal Transformers across SAR and Multispectral Modalities
• GeLoc3r: Enhancing Relative Camera Pose Regression with Geometric Consistency Regularization
• MMeViT: Multi-Modal ensemble ViT for Post-Stroke Rehabilitation Action Recognition
• Activation Matching for Explanation Generation
• InfoDet: A Dataset for Infographic Element Detection
• On the Design of KL-Regularized Policy Gradient Algorithms for LLM Reasoning
• Trinity-RFT: A General-Purpose and Unified Framework for Reinforcement Fine-Tuning of Large Language Models
• Reward Model Overoptimisation in Iterated RLHF
• TabularGSM: Understanding the Limitations of LLMs in Tabular Math Reasoning
• HS-STaR: Hierarchical Sampling for Self-Taught Reasoners via Difficulty Estimation and Budget Reallocation
• Cross-modal RAG: Sub-dimensional Text-to-Image Retrieval-Augmented Generation
• ProxyThinker: Test-Time Guidance through Small Visual Reasoners
• Comba: Improving Bilinear RNNs with Closed-loop Control
• VisCoder: Fine-Tuning LLMs for Executable Python Visualization Code Generation
• InstructPro: Natural Language Guided Ligand-Binding Protein Design
• One Patient, Many Contexts: Scaling Medical AI with Contextual Intelligence
• Beyond Jailbreaking: Auditing Contextual Privacy in LLM Agents
• Discrete Audio Tokens: More Than a Survey!
• Enhancing Delta Compression in LLMs via SVD-based Quantization Error Minimization
• OmniGen2: Exploration to Advanced Multimodal Generation
• Overcoming Long-Context Limitations of State-Space Models via Context-Dependent Sparse Attention
• One Token to Fool LLM-as-a-Judge
• MetaLint: Generalizable Idiomatic Code Quality Analysis through Instruction-Following and Easy-to-Hard Generalization
• Probabilistic Soundness Guarantees in LLM Reasoning Chains
• A Markov Categorical Framework for Language Modeling
• Can Language Models Discover Scaling Laws?
• CADDesigner: Conceptual Design of CAD Models Based on General-Purpose Agent
• Trainable Dynamic Mask Sparse Attention
• AttriLens-Mol: Attribute Guided Reinforcement Learning for Molecular Property Prediction with Large Language Models
• Attention Layers Add Into Low-Dimensional Residual Subspaces
• Evaluating the Effectiveness of Transformer Layers in Wav2Vec 2.0, XLS-R, and Whisper for Speaker Identification Tasks
• Emergent Hierarchical Reasoning in LLMs through Reinforcement Learning
• Hierarchical Task Environments as the Next Frontier for Embodied World Models in Robot Soccer
• FuseCodec: Semantic-Contextual Fusion and Supervision for Neural Codecs
• TDRM: Smooth Reward Models with Temporal Difference for LLM RL and Inference
• Patterns in the Transition From Founder-Leadership to Community Governance of Open Source
• PiERN: Token-Level Routing for Integrating High-Precision Computation and Reasoning
• The Thinking Spectrum: An Empirical Study of Tunable Reasoning in LLMs through Model Merging
• Reasoning Isn't Enough: Examining Truth-Bias and Sycophancy in LLMs
• Don't Trust Generative Agents to Mimic Communication on Social Networks Unless You Benchmarked their Empirical Realism
• Semantic-guided Diverse Decoding for Large Language Model
• Latent Chain-of-Thought? Decoding the Depth-Recurrent Transformer
• PRIME: Large Language Model Personalization with Cognitive Dual-Memory and Personalized Thought Process
• CoSteer: Collaborative Decoding-Time Personalization via Local Delta Steering
• ArtifactsBench: Bridging the Visual-Interactive Gap in LLM Code Generation Evaluation
• Entropy-Memorization Law: Evaluating Memorization Difficulty of Data in LLMs
• Function Induction and Task Generalization: An Interpretability Study with Off-by-One Addition
• Making Language Model a Hierarchical Classifier
• LionGuard 2: Building Lightweight, Data-Efficient & Localised Multilingual Content Moderators
• The Ever-Evolving Science Exam
• CTTS: Collective Test-Time Scaling
• Discerning minds or generic tutors? Evaluating instructional guidance capabilities in Socratic LLMs
• Coarse-to-Fine Personalized LLM Impressions for Streamlined Radiology Reports
• Jet-Nemotron: Efficient Language Model with Post Neural Architecture Search
• CORE-RAG: Lossless Compression for Retrieval-Augmented LLMs via Reinforcement Learning
• Automatic Question & Answer Generation Using Generative Large Language Model (LLM)
• When Thinking Backfires: Mechanistic Insights Into Reasoning-Induced Misalignment
• Causal Attention with Lookahead Keys
• ATTS: Asynchronous Test-Time Scaling via Conformal Prediction
• Fast and Fluent Diffusion Language Models via Convolutional Decoding and Rejective Fine-tuning
• Agentic Reinforcement Learning with Implicit Step Rewards
• Towards Efficient LLM Grounding for Embodied Multi-Agent Collaboration
• Position: Towards Bidirectional Human-AI Alignment
• A Voter-Based Stochastic Rejection-Method Framework for Asymptotically Safe Language Model Outputs
• NextLocLLM: Location Semantics Modeling and Coordinate-Based Next Location Prediction with LLMs
• Similarity-Dissimilarity Loss for Multi-label Supervised Contrastive Learning
• CoT-TL: Low-Resource Temporal Knowledge Representation of Planning Instructions Using Chain-of-Thought Reasoning
• A Neurosymbolic Fast and Slow Architecture for Graph Coloring
• Sometimes I am a Tree: Data Drives Unstable Hierarchical Generalization
• Collaborative Gym: A Framework for Enabling and Evaluating Human-Agent Collaboration
• Reversed in Time: A Novel Temporal-Emphasized Benchmark for Cross-Modal Video-Text Retrieval
• Mind the Value-Action Gap: Do LLMs Act in Alignment with Their Values?
• vCache: Verified Semantic Prompt Caching
• SURGE: On the Potential of Large Language Models as General-Purpose Surrogate Code Executors
• Mitigating Barren Plateaus in Quantum Neural Networks via an AI-Driven Submartingale-Based Framework
• Reasoning to Learn from Latent Thoughts
• MaintainCoder: Maintainable Code Generation Under Dynamic Requirements
• Do Larger Language Models Generalize Better? A Scaling Law for Implicit Reasoning at Pretraining Time
• Visual Planning: Let's Think Only with Images
• Signal in the Noise: Polysemantic Interference Transfers and Predicts Cross-Model Influence
• AgentThink: A Unified Framework for Tool-Augmented Chain-of-Thought Reasoning in Vision-Language Models for Autonomous Driving
• Better Safe Than Sorry? Overreaction Problem of Vision Language Models in Visual Emergency Recognition
• OViP: Online Vision-Language Preference Learning for VLM Hallucination
• AdaSTaR: Adaptive Data Sampling for Training Self-Taught Reasoners
• DEL-ToM: Inference-Time Scaling for Theory-of-Mind Reasoning via Dynamic Epistemic Logic
• LLMs Are In-Context Bandit Reinforcement Learners
• AERA Chat: An Interactive Platform for Automated Explainable Student Answer Assessment
• DM-Codec: Distilling Multimodal Representations for Speech Tokenization
• When Speculation Spills Secrets: Side Channels via Speculative Decoding In LLMs
• Adapting Chat Language Models Using Only Target Unlabeled Language Data
• LongProc: Benchmarking Long-Context Language Models on Long Procedural Generation
• A Partition Cover Approach to Tokenization
• A Survey of Graph Retrieval-Augmented Generation for Customized Large Language Models
• ESGSenticNet: A Neurosymbolic Knowledge Base for Corporate Sustainability Analysis
• Beyond checkmate: exploring the creative chokepoints in AI text
• Which Words Matter Most in Zero-Shot Prompts?
• UltraIF: Advancing Instruction Following from the Wild
• Group-Adaptive Threshold Optimization for Robust AI-Generated Text Detection
• Confidence Improves Self-Consistency in LLMs
• PAFT: Prompt-Agnostic Fine-Tuning
• B-cos LM: Efficiently Transforming Pre-trained Language Models for Improved Explainability
• PropXplain: Can LLMs Enable Explainable Propaganda Detection?
• MemeIntel: Explainable Detection of Propagandistic and Hateful Memes
• Two Heads Are Better Than One: Dual-Model Verbal Reflection at Inference-Time
• How to Protect Yourself from 5G Radiation? Investigating LLM Responses to Implicit Misinformation
• Adaptive Group Policy Optimization: Towards Stable Training and Token-Efficient Reasoning
• SUV: Scalable Large Language Model Copyright Compliance with Regularized Selective Unlearning
• XL-Suite: Cross-Lingual Synthetic Training and Evaluation Data for Open-Ended Generation
• SentenceKV: Efficient LLM Inference via Sentence-Level Semantic KV Caching
• AnesSuite: A Comprehensive Benchmark and Dataset Suite for Anesthesiology Reasoning in LLMs
• A Practical Synthesis of Detecting AI-Generated Textual, Visual, and Audio Content
• Are You Getting What You Pay For? Auditing Model Substitution in LLM APIs
• DataPuzzle: Breaking Free from the Hallucinated Promise of LLMs in Data Analysis
• Efficient Reasoning Models: A Survey
• IPBench: Benchmarking the Knowledge of Large Language Models in Intellectual Property
• Dynamic Early Exit in Reasoning Models
• TRACE Back from the Future: A Probabilistic Reasoning Approach to Controllable Language Generation
• Cooking Up Creativity: Enhancing LLM Creativity through Structured Recombination
• $\textit{New News}$: System-2 Fine-tuning for Robust Integration of New Knowledge
• References Indeed Matter? Reference-Free Preference Optimization for Conversational Query Reformulation
• OnPrem.LLM: A Privacy-Conscious Document Intelligence Toolkit
• VeriFact: Enhancing Long-Form Factuality Evaluation with Refined Fact Extraction and Reference Facts
• Learning When to Think: Shaping Adaptive Reasoning in R1-Style Models via Multi-Stage RL
• The Counting Power of Transformers
• Critique-Guided Distillation for Efficient and Robust Language Model Reasoning
• AdaBoN: Adaptive Best-of-N Alignment
• MobileIPL: Enhancing Mobile Agents Thinking Process via Iterative Preference Learning
• Automatically Advancing LLM Expertise in Technology Judgment
• Is Active Persona Inference Necessary for Aligning Small Models to Personal Preferences?
• Game-RL: Synthesizing Multimodal Verifiable Game Data to Boost VLMs' General Reasoning
• Mechanistic Fine-tuning for In-context Learning
• Internal Chain-of-Thought: Empirical Evidence for Layer-wise Subtask Scheduling in LLMs
• Language Models Optimized to Fool Detectors Still Have a Distinct Style (And How to Change It)
• ReflAct: World-Grounded Decision Making in LLM Agents via Goal-State Reflection
• Multilingual Prompting for Improving LLM Generation Diversity
• Generalizable Process Reward Models via Formally Verified Training Data
• Leveraging Online Data to Enhance Medical Knowledge in a Small Persian Language Model
• Align-GRAG: Reasoning-Guided Dual Alignment for Graph Retrieval-Augmented Generation
• ToDi: Token-wise Distillation via Fine-Grained Divergence Control
• Nested Named Entity Recognition as Single-Pass Sequence Labeling
• A Survey on Stereotype Detection in Natural Language Processing
• BRIT: Bidirectional Retrieval over Unified Image-Text Graph
• MetaMind: Modeling Human Social Thoughts with Metacognitive Multi-Agent Systems
• A Necessary Step toward Faithfulness: Measuring and Improving Consistency in Free-Text Explanations
• From Single to Multi-Granularity: Toward Long-Term Memory Association and Selection of Conversational Agents
• TrojanStego: Your Language Model Can Secretly Be A Steganographic Privacy Leaking Agent
• Long Context Scaling: Divide and Conquer via Multi-Agent Question-driven Collaboration
• SpecExtend: A Drop-in Enhancement for Speculative Decoding of Long Sequences
• Evaluating and Steering Modality Preferences in Multimodal Large Language Model
• Pearl: A Multimodal Culturally-Aware Arabic Instruction Dataset
• Semi-structured LLM Reasoners Can Be Rigorously Audited
• Improving Reliability and Explainability of Medical Question Answering through Atomic Fact Checking in Retrieval-Augmented LLMs
• Hanfu-Bench: A Multimodal Benchmark on Cross-Temporal Cultural Understanding and Transcreation
• Answer Convergence as a Signal for Early Stopping in Reasoning
• Beyond Classification: Towards Speech Emotion Reasoning with Multitask AudioLLMs
• Improving LLM Reasoning through Interpretable Role-Playing Steering
• What Do Indonesians Really Need from Language Technology? A Nationwide Survey
• AdversariaL attacK sAfety aLIgnment(ALKALI): Safeguarding LLMs through GRACE: Geometric Representation-Aware Contrastive Enhancement- Introducing Adversarial Vulnerability Quality Index (AVQI)
• Query-Focused Retrieval Heads Improve Long-Context Reasoning and Re-ranking
• Curriculum-Guided Layer Scaling for Language Model Pretraining
• Infini-gram mini: Exact n-gram Search at the Internet Scale with FM-Index
• BOW: Reinforcement Learning for Bottlenecked Next Word Prediction
• Long-Context Generalization with Sparse Attention
• GRAF: Multi-turn Jailbreaking via Global Refinement and Active Fabrication
• Beyond Isolated Facts: Synthesizing Narrative and Grounded Supervision for VideoQA
• Euclid's Gift: Enhancing Spatial Perception and Reasoning in Vision-Language Models via Geometric Surrogate Tasks
• Experience-guided reflective co-evolution of prompts and heuristics for automatic algorithm design
• LEAF: A Robust Expert-Based Framework for Few-Shot Continual Event Detection
• NeMo: Needle in a Montage for Video-Language Understanding
• OrthAlign: Orthogonal Subspace Decomposition for Non-Interfering Multi-Objective Alignment
• On the Self-awareness of Large Reasoning Models' Capability Boundaries
• VSSFlow: Unifying Video-conditioned Sound and Speech Generation via Joint Learning
• Pushing LLMs to Their Logical Reasoning Bound: The Role of Data Reasoning Intensity
• Retro*: Optimizing LLMs for Reasoning-Intensive Document Retrieval
• MMRQA: Signal-Enhanced Multimodal Large Language Models for MRI Quality Assessment
• Neural network embeddings recover value dimensions from psychometric survey items on par with human data
• MASLegalBench: Benchmarking Multi-Agent Systems in Deductive Legal Reasoning
• When Greedy Wins: Emergent Exploitation Bias in Meta-Bandit LLM Training
• DiffTester: Accelerating Unit Test Generation for Diffusion LLMs via Repetitive Pattern
• Learning from Convenience Samples: A Case Study on Fine-Tuning LLMs for Survey Non-response in the German Longitudinal Election Study
• Scaling with Collapse: Efficient and Predictable Training of LLM Families
• ORPO-Distill: Mixed-Policy Preference Optimization for Cross-Architecture LLM Distillation
• From $f(x)$ and $g(x)$ to $f(g(x))$: LLMs Learn New Skills in RL by Composing Old Ones
• MGM-Omni: Scaling Omni LLMs to Personalized Long-Horizon Speech
• Rethinking Entropy Regularization in Large Reasoning Models
• The Era of Real-World Human Interaction: RL from User Conversations
• ReasoningBank: Scaling Agent Self-Evolving with Reasoning Memory
• TemMed-Bench: Evaluating Temporal Medical Image Reasoning in Vision-Language Models
• GSM8K-V: Can Vision Language Models Solve Grade School Math Word Problems in Visual Contexts
• SIRI: Scaling Iterative Reinforcement Learning with Interleaved Compression
• WordAlchemy: A transformer-based Reverse Dictionary
• Continual Dialogue State Tracking via Example-Guided Question Answering
• CGELBank Annotation Manual v1.2
• Machines Do See Color: A Guideline to Classify Different Forms of Racist Discourse in Large Corpora
• Enhancing Textual Personality Detection toward Social Media: Integrating Long-term and Short-term Perspectives
• Multi-Head RAG: Solving Multi-Aspect Problems with LLMs
• SciRIFF: A Resource to Enhance Language Model Instruction-Following over Scientific Literature
• Sheaf Discovery with Joint Computation Graph Pruning and Flexible Granularity
• CiteFusion: An Ensemble Framework for Citation Intent Classification Harnessing Dual-Model Binary Couples and SHAP Analyses
• LLM-3D Print: Large Language Models To Monitor and Control 3D Printing
• Cross-lingual Human-Preference Alignment for Neural Machine Translation with Direct Quality Optimization
• Your Models Have Thought Enough: Training Large Reasoning Models to Stop Overthinking
• SPIKE-RL: Video-LLMs meet Bayesian Surprise
• FoR-SALE: Frame of Reference-guided Spatial Adjustment in LLM-based Diffusion Editing
• MaskSQL: Safeguarding Privacy for LLM-Based Text-to-SQL via Abstraction
• Temporal Generalization: A Reality Check
• Mapping Overlaps in Benchmarks through Perplexity in the Wild
• Multi-modal Data Spectrum: Multi-modal Datasets are Multi-dimensional
• Clean First, Align Later: Benchmarking Preference Data Cleaning for Reliable LLM Alignment
• RIV: Recursive Introspection Mask Diffusion Vision Language Model
• From Past To Path: Masked History Learning for Next-Item Prediction in Generative Recommendation
• RCI: A Score for Evaluating Global and Local Reasoning in Multimodal Benchmarks
• From Reasoning to Answer: Empirical, Attention-Based and Mechanistic Insights into Distilled DeepSeek R1 Models
• Towards a Comprehensive Scaling Law of Mixture-of-Experts
• HomeSafeBench: A Benchmark for Embodied Vision-Language Models in Free-Exploration Home Safety Inspection
• SafeSearch: Automated Red-Teaming for the Safety of LLM-Based Search Agents
• Beyond Game Theory Optimal: Profit-Maximizing Poker Agents for No-Limit Holdem
• Anchored Supervised Fine-Tuning
• From What to Why: A Multi-Agent System for Evidence-based Chemical Reaction Condition Reasoning
• Knowledge Homophily in Large Language Models
• Beyond the Exploration-Exploitation Trade-off: A Hidden State Approach for LLM Reasoning in RLVR
• PCRI: Measuring Context Robustness in Multimodal Models for Enterprise Applications
• Dynamic Orthogonal Continual Fine-tuning for Mitigating Catastrophic Forgettings
• Beyond Benchmarks: Understanding Mixture-of-Experts Models through Internal Mechanisms
• Explore-Execute Chain: Towards an Efficient Structured Reasoning Paradigm
• Conditional Advantage Estimation for Reinforcement Learning in Large Reasoning Models
• Detecting and Rectifying Noisy Labels: A Similarity-based Approach
• The Role of Logic and Automata in Understanding Transformers
• Do Repetitions Matter? Strengthening Reliability in LLM Evaluations
• Generalist Scanner Meets Specialist Locator: A Synergistic Coarse-to-Fine Framework for Robust GUI Grounding
• Reasoning or Retrieval? A Study of Answer Attribution on Large Reasoning Models
• Group-Relative REINFORCE Is Secretly an Off-Policy Algorithm: Demystifying Some Myths About GRPO and Its Friends
• Metamorphic Testing for Audio Content Moderation Software
• Learning to Ponder: Adaptive Reasoning in Latent Space
• SpecExit: Accelerating Large Reasoning Model via Speculative Exit
• Latent Visual Reasoning
• Extracting the Structure of Press Releases for Predicting Earnings Announcement Returns
• PAME-AI: Patient Messaging Creation and Optimization using Agentic AI
• AdvChain: Adversarial Chain-of-Thought Tuning for Robust Safety Alignment of Large Reasoning Models
• Overview of SCIDOCA 2025 Shared Task on Citation Prediction, Discovery, and Placement
• SCI-Verifier: Scientific Verifier with Thinking
• Bridging the behavior-neural gap: A multimodal AI reveals the brain's geometry of emotion more accurately than human self-reports
• MAS$^2$: Self-Generative, Self-Configuring, Self-Rectifying Multi-Agent Systems
• Towards Safe Reasoning in Large Reasoning Models via Corrective Intervention
• Beyond Repetition: Text Simplification and Curriculum Learning for Data-Constrained Pretraining
• Reinforcement Mid-Training
• HarmMetric Eval: Benchmarking Metrics and Judges for LLM Harmfulness Assessment
• LLaDA-MoE: A Sparse MoE Diffusion Language Model
• Agentar-Scale-SQL: Advancing Text-to-SQL through Orchestrated Test-Time Scaling
• Multilingual Text-to-SQL: Benchmarking the Limits of Language Models with Collaborative Language Agents
• CDT: A Comprehensive Capability Framework for Large Language Models Across Cognition, Domain, and Task
• Alternatives To Next Token Prediction In Text Generation - A Survey
• Bias Mitigation or Cultural Commonsense? Evaluating LLMs with a Japanese Dataset
• A Text-To-Text Alignment Algorithm for Better Evaluation of Modern Speech Recognition Systems
• Sanitize Your Responses: Mitigating Privacy Leakage in Large Language Models
• GRPO-MA: Multi-Answer Generation in GRPO for Stable and Efficient Chain-of-Thought Training
• Knowledge Editing with Subspace-Aware Key-Value Mappings
• Building Benchmarks from the Ground Up: Community-Centered Evaluation of LLMs in Healthcare Chatbot Settings
• AdaThink-Med: Medical Adaptive Thinking with Uncertainty-Guided Length Calibration
• Inducing Dyslexia in Vision Language Models
• HiKE: Hierarchical Evaluation Framework for Korean-English Code-Switching Speech Recognition
• Hype or not? Formalizing Automatic Promotional Language Detection in Biomedical Research
• InfLLM-V2: Dense-Sparse Switchable Attention for Seamless Short-to-Long Adaptation
• Understanding the Dilemma of Unlearning for Large Language Models
• Reference-Free Rating of LLM Responses via Latent Information
• MemGen: Weaving Generative Latent Memory for Self-Evolving Agents
• Socratic-Zero : Bootstrapping Reasoning via Data-Free Agent Co-evolution
• ProxyAttn: Guided Sparse Attention via Representative Heads
• LatentEvolve: Self-Evolving Test-Time Scaling in Latent Space
• SeaPO: Strategic Error Amplification for Robust Preference Optimization of Large Language Models
• Evaluating Spatiotemporal Consistency in Automatically Generated Sewing Instructions
• KnowGuard: Knowledge-Driven Abstention for Multi-Round Clinical Reasoning
• DiaCDM: Cognitive Diagnosis in Teacher-Student Dialogues using the Initiation-Response-Evaluation Framework
• SemShareKV: Efficient KVCache Sharing for Semantically Similar Prompts via Token-Level LSH Matching
• Hierarchical Error Correction for Large Language Models: A Systematic Framework for Domain-Specific AI Quality Enhancement
• Between Help and Harm: An Evaluation of Mental Health Crisis Handling by LLMs
• Metaphor identification using large language models: A comparison of RAG, prompt engineering, and fine-tuning
• Expanding Computation Spaces of LLMs at Inference Time
• BOE-XSUM: Extreme Summarization in Clear Language of Spanish Legal Decrees and Notifications
• How Well Do LLMs Imitate Human Writing Style?
• MobileLLM-R1: Exploring the Limits of Sub-Billion Language Model Reasoners with Open Training Recipes
• The Dialogue That Heals: A Comprehensive Evaluation of Doctor Agents' Inquiry Capability
• SemanticShield: LLM-Powered Audits Expose Shilling Attacks in Recommender Systems
• Generalized Correctness Models: Learning Calibrated and Model-Agnostic Correctness Predictors from Historical Patterns
• Circuit Distillation
• Ultra-Fast Language Generation via Discrete Diffusion Divergence Instruct
• GateMABSA: Aspect-Image Gated Fusion for Multimodal Aspect-based Sentiment Analysis
• Hyperdimensional Probe: Decoding LLM Representations via Vector Symbolic Architectures
• Confidence-Guided Error Correction for Disordered Speech Recognition
• An empirical study on the limitation of Transformers in program trace generation
• Scaling Generalist Data-Analytic Agents
• jina-reranker-v3: Last but Not Late Interaction for Document Reranking
• Towards Trustworthy Lexical Simplification: Exploring Safety and Efficiency with Small LLMs
• Towards Personalized Deep Research: Benchmarks and Evaluations
• Knowledge Extraction on Semi-Structured Content: Does It Remain Relevant for Question Answering in the Era of LLMs?
• Investigating Language and Retrieval Bias in Multilingual Previously Fact-Checked Claim Detection
• Paired by the Teacher: Turning Unpaired Data into High-Fidelity Pairs for Low-Resource Text Generation
• Pretraining Large Language Models with NVFP4
• EasySteer: A Unified Framework for High-Performance and Extensible LLM Steering
• NAIPv2: Debiased Pairwise Learning for Efficient Paper Quality Estimation
• Incentive-Aligned Multi-Source LLM Summaries
• Learning to Parallel: Accelerating Diffusion Large Language Models via Adaptive Parallel Decoding
• InfoAgent: Advancing Autonomous Information-Seeking Agents
• CAOTE: KV Cache Selection for LLMs via Attention Output Error-Based Token Eviction
• Multiplicative-Additive Constrained Models:Toward Joint Visualization of Interactive and Independent Effects
• Benefits and Pitfalls of Reinforcement Learning for Language Model Planning: A Theoretical Perspective
• DiaMoE-TTS: A Unified IPA-Based Dialect TTS Framework with Mixture-of-Experts and Parameter-Efficient Zero-Shot Adaptation
• VideoScore2: Think before You Score in Generative Video Evaluation
• Toward a Theory of Generalizability in LLM Mechanistic Interpretability Research
• Adaptive Margin RLHF via Preference over Preferences
• Patient-specific Biomolecular Instruction Tuning
• JE-IRT: A Geometric Lens on LLM Abilities through Joint Embedding Item Response Theory
• Not only a helper, but also a teacher: Interactive LLM Cascade
• Geometry-Aware Losses for Structure-Preserving Text-to-Sign Language Generation
• Tracing the Representation Geometry of Language Models from Pretraining to Post-training
• Virus Infection Attack on LLMs: Your Poisoning Can Spread "VIA" Synthetic Data
• Kimi-Dev: Agentless Training as Skill Prior for SWE-Agents
• Causally-Enhanced Reinforcement Policy Optimization
• Multiplayer Nash Preference Optimization
• RHYTHM: Reasoning with Hierarchical Temporal Tokenization for Human Mobility
• C$^2$GSPG: Confidence-calibrated Group Sequence Policy Gradient towards Self-aware Reasoning
• SPEC-RL: Accelerating On-Policy Reinforcement Learning via Speculative Rollouts
• $p$-less Sampling: A Robust Hyperparameter-Free Approach for LLM Decoding
• Learning How to Use Tools, Not Just When: Pattern-Aware Tool-Integrated Reasoning
• Seeing Symbols, Missing Cultures: Probing Vision-Language Models' Reasoning on Fire Imagery and Cultural Meaning
• PARROT: A Benchmark for Evaluating LLMs in Cross-System SQL Translation
• Guard Vector: Beyond English LLM Guardrails with Task-Vector Composition and Streaming-Aware Prefix SFT
• Train Once, Answer All: Many Pretraining Experiments for the Cost of One
• No Loss, No Gain: Gated Refinement and Adaptive Compression for Prompt Optimization
• Liaozhai through the Looking-Glass: On Paratextual Explicitation of Culture-Bound Terms in Machine Translation
• Comparison of Scoring Rationales Between Large Language Models and Human Raters
• Retrieval-Constrained Decoding Reveals Underestimated Parametric Knowledge in Language Models
• Cognition-of-Thought Elicits Social-Aligned Reasoning in Large Language Models
• Text-Based Approaches to Item Difficulty Modeling in Large-Scale Assessments: A Systematic Review
• The Impact of Role Design in In-Context Learning for Large Language Models
• AraS2P: Arabic Speech-to-Phonemes System
• From Human Annotation to Automation: LLM-in-the-Loop Active Learning for Arabic Sentiment Analysis
• On the Shelf Life of Fine-Tuned LLM Judges: Future Proofing, Backward Compatibility, and Question Generalization
• Automatic Speech Recognition for Greek Medical Dictation
• Towards Efficient CoT Distillation: Self-Guided Rationale Selector for Better Performance with Fewer Rationales
• Jackal: A Real-World Execution-Based Benchmark Evaluating Large Language Models on Text-to-JQL Tasks
• LLM Hallucination Detection: HSAD
• Timber: Training-free Instruct Model Refining with Base via Effective Rank
• Fast Thinking for Large Language Models
• Don't Settle Too Early: Self-Reflective Remasking for Diffusion Language Models
• Beyond English-Centric Training: How Reinforcement Learning Improves Cross-Lingual Reasoning in LLMs
• Aligning LLMs for Multilingual Consistency in Enterprise Applications
• TF-Bench: Evaluating Program Semantics Reasoning with Type Inference in System F
• VIVA+: Human-Centered Situational Decision-Making
• Collaboration of Fusion and Independence: Hypercomplex-driven Robust Multi-Modal Knowledge Graph Completion
• Do LLMs Understand Romanian Driving Laws? A Study on Multimodal and Fine-Tuned Question Answering
• Compose and Fuse: Revisiting the Foundational Bottlenecks in Multimodal Reasoning
• Understanding Textual Capability Degradation in Speech LLMs via Parameter Importance Analysis
• Knowledge-Level Consistency Reinforcement Learning: Dual-Fact Alignment for Long-Form Factuality
• From Personal to Collective: On the Role of Local and Global Memory in LLM Personalization
• Bridging the Knowledge-Prediction Gap in LLMs on Multiple-Choice Questions
• Transformer Tafsir at QIAS 2025 Shared Task: Hybrid Retrieval-Augmented Generation for Islamic Knowledge Question Answering
• Open-DeBias: Toward Mitigating Open-Set Bias in Language Models
• SPELL: Self-Play Reinforcement Learning for evolving Long-Context Language Models
• Winning the Pruning Gamble: A Unified Approach to Joint Sample and Token Pruning for Efficient Supervised Fine-Tuning
• DocPruner: A Storage-Efficient Framework for Multi-Vector Visual Document Retrieval via Adaptive Patch-Level Embedding Pruning
• Taming Masked Diffusion Language Models via Consistency Trajectory Reinforcement Learning with Fewer Decoding Step
• Assessing Large Language Models in Updating Their Forecasts with New Information
• Easy Turn: Integrating Acoustic and Linguistic Modalities for Robust Turn-Taking in Full-Duplex Spoken Dialogue Systems
• Vision-Grounded Machine Interpreting: Improving the Translation Process through Visual Cues
• HiPO: Hybrid Policy Optimization for Dynamic Reasoning in LLMs
• ByteSized32Refactored: Towards an Extensible Interactive Text Games Corpus for LLM World Modeling and Evaluation
• Toward Preference-aligned Large Language Models via Residual-based Model Steering
• The Hidden Costs of Translation Accuracy: Distillation, Quantization, and Environmental Impact
• The AI Agent Code of Conduct: Automated Guardrail Policy-as-Prompt Synthesis
• MCPMark: A Benchmark for Stress-Testing Realistic and Comprehensive MCP Use
• Sequential Diffusion Language Models
• SparseD: Sparse Attention for Diffusion Language Models
• ResFormer: All-Time Reservoir Memory for Long Sequence Classification
• Ensembling Multilingual Transformers for Robust Sentiment Analysis of Tweets
• Large-Scale Constraint Generation - Can LLMs Parse Hundreds of Constraints?
• GEAR: A General Evaluation Framework for Abductive Reasoning
• BTC-SAM: Leveraging LLMs for Generation of Bias Test Cases for Sentiment Analysis Models
• Pragmatic Inference for Moral Reasoning Acquisition: Generalization via Distributional Semantics
• Dual-Scale World Models for LLM Agents Towards Hard-Exploration Problems
• EduVidQA: Generating and Evaluating Long-form Answers to Student Questions based on Lecture Videos
• Beyond Magic Words: Sharpness-Aware Prompt Evolving for Robust Large Language Models with TARE
• Your thoughts tell who you are: Characterize the reasoning patterns of LRMs
• Localizing Task Recognition and Task Learning in In-Context Learning via Attention Head Analysis
• Task Vectors, Learned Not Extracted: Performance Gains and Mechanistic Insight
• Retrieval-augmented GUI Agents with Generative Guidelines
• Beyond Overall Accuracy: A Psychometric Deep Dive into the Topic-Specific Medical Capabilities of 80 Large Language Models
• PET: Preference Evolution Tracking with LLM-Generated Explainable Distribution
• AceSearcher: Bootstrapping Reasoning and Search for LLMs via Reinforced Self-Play
• Can Large Language Models Express Uncertainty Like Human?
• BeyondBench: Benchmark-Free Evaluation of Reasoning in Language Models
• ScenarioBench: Trace-Grounded Compliance Evaluation for Text-to-SQL and RAG
• MoVa: Towards Generalizable Classification of Human Morals and Values
• Model Fusion with Multi-LoRA Inference for Tool-Enhanced Game Dialogue Agents
• Prompt and Parameter Co-Optimization for Large Language Models
• MRAG-Suite: A Diagnostic Evaluation Platform for Visual Retrieval-Augmented Generation
• SimuHome: A Temporal- and Environment-Aware Benchmark for Smart Home LLM Agents
• Let LLMs Speak Embedding Languages: Generative Text Embeddings via Iterative Contrastive Refinement
• LOGOS: LLM-driven End-to-End Grounded Theory Development and Schema Induction for Qualitative Research
• DiffuGuard: How Intrinsic Safety is Lost and Found in Diffusion Large Language Models
• Q-Mirror: Unlocking the Multi-Modal Potential of Scientific Text-Only QA Pairs
• Dual Mechanisms of Value Expression: Intrinsic vs. Prompted Values in LLMs
• Multimodal Large Language Models Meet Multimodal Emotion Recognition and Reasoning: A Survey
• Speculative Verification: Exploiting Information Gain to Refine Speculative Decoding
• AlignX: Advancing Multilingual Large Language Models with Multilingual Representation Alignment
• Multi-Modal Sentiment Analysis with Dynamic Attention Fusion
• Enabling Approximate Joint Sampling in Diffusion LMs
• Painless Activation Steering: An Automated, Lightweight Approach for Post-Training Large Language Models
• MIRAGE: Multi-hop Reasoning with Ambiguity Evaluation for Illusory Questions
• ML2B: Multi-Lingual ML Benchmark For AutoML
• ArFake: A Multi-Dialect Benchmark and Baselines for Arabic Spoof-Speech Detection
• EditGRPO: Reinforcement Learning with Post -Rollout Edits for Clinically Accurate Chest X-Ray Report Generation
• Critique-Coder: Enhancing Coder Models by Critique Reinforcement Learning
• ChatInject: Abusing Chat Templates for Prompt Injection in LLM Agents
• Learning to Detect Relevant Contexts and Knowledge for Response Selection in Retrieval-based Dialogue Systems
• Towards Generalizable Implicit In-Context Learning with Attention Routing
• The Bias is in the Details: An Assessment of Cognitive Bias in LLMs
• Lexicon-Enriched Graph Modeling for Arabic Document Readability Prediction
• HEART: Emotionally-driven test-time scaling of Language Models
• Infusing Theory of Mind into Socially Intelligent LLM Agents
• Extract-0: A Specialized Language Model for Document Information Extraction
• Large language models management of medications: three performance analyses
• LLMs Behind the Scenes: Enabling Narrative Scene Illustration
• What Matters More For In-Context Learning under Matched Compute Budgets: Pretraining on Natural Text or Incorporating Targeted Synthetic Examples?
• Emergent morpho-phonological representations in self-supervised speech models
• Same Content, Different Representations: A Controlled Study for Table QA
• ADAM: A Diverse Archive of Mankind for Evaluating and Enhancing LLMs in Biographical Reasoning
• AI Brown and AI Koditex: LLM-Generated Corpora Comparable to Traditional Corpora of English and Czech Texts
• Look Back to Reason Forward: Revisitable Memory for Long-Context LLM Agents
• Peacemaker or Troublemaker: How Sycophancy Shapes Multi-Agent Debate
• Semantic Voting: A Self-Evaluation-Free Approach for Efficient LLM Self-Improvement on Unverifiable Open-ended Tasks
• From Evidence to Trajectory: Abductive Reasoning Path Synthesis for Training Retrieval-Augmented Generation Agents
• The Geometry of Creative Variability: How Credal Sets Expose Calibration Gaps in Language Models
• d$^2$Cache: Accelerating Diffusion-Based LLMs via Dual Adaptive Caching
• How to Make Large Language Models Generate 100% Valid Molecules?
• Non-Collaborative User Simulators for Tool Agents
• Tagging the Thought: Unlocking Personalization Reasoning via Reinforcement Learning
• Tree Reward-Aligned Search for TReASURe in Masked Diffusion Language Models
• Test-Time Policy Adaptation for Enhanced Multi-Turn Interactions with LLMs
• Pretraining LLM with Latent Thoughts in Continuous Space
• Diagnose, Localize, Align: A Full-Stack Framework for Reliable LLM Multi-Agent Systems under Instruction Conflicts
• Estimating the strength and timing of syntactic structure building in naturalistic reading
• From Harm to Help: Turning Reasoning In-Context Demos into Assets for Reasoning LMs
• Global Beats, Local Tongue: Studying Code Switching in K-pop Hits on Billboard Charts
• Steering Prepositional Phrases in Language Models: A Case of with-headed Adjectival and Adverbial Complements in Gemma-2
• PARL-MT: Learning to Call Functions in Multi-Turn Conversation with Progress Awareness
• A Structured Framework for Evaluating and Enhancing Interpretive Capabilities of Multimodal LLMs in Culturally Situated Tasks
• Detecting Corpus-Level Knowledge Inconsistencies in Wikipedia with Large Language Models
• Fin-ExBERT: User Intent based Text Extraction in Financial Context using Graph-Augmented BERT and trainable Plugin
• A2D: Any-Order, Any-Step Safety Alignment for Diffusion Language Models
• Scaling Policy Compliance Assessment in Language Models with Policy Reasoning Traces
• Learning to Reason in Structured In-context Environments with Reinforcement Learning
• C-Evolve: Consensus-based Evolution for Prompt Groups
• Dual-Space Smoothness for Robust and Balanced LLM Unlearning
• MedCritical: Enhancing Medical Reasoning in Small Language Models via Self-Collaborative Correction
• Alignment through Meta-Weighted Online Sampling: Bridging the Gap between Data Generation and Preference Optimization
• CCD: Mitigating Hallucinations in Radiology MLLMs via Clinical Contrastive Decoding
• Are you sure? Measuring models bias in content moderation through uncertainty
• AccessEval: Benchmarking Disability Bias in Large Language Models
• RAR$^2$: Retrieval-Augmented Medical Reasoning via Thought-Driven Retrieval
• TRUEBench: Can LLM Response Meet Real-world Constraints as Productivity Assistant?
• Comprehension Without Competence: Architectural Limits of LLMs in Symbolic Computation and Reasoning
• DuetGraph: Coarse-to-Fine Knowledge Graph Reasoning with Dual-Pathway Global-Local Fusion
• MetaLint: Generalizable Idiomatic Code Quality Analysis through Instruction-Following and Easy-to-Hard Generalization
• BenchRL-QAS: Benchmarking reinforcement learning algorithms for quantum architecture search
• LionGuard 2: Building Lightweight, Data-Efficient & Localised Multilingual Content Moderators
• Diffusion models for multivariate subsurface generation and efficient probabilistic inversion
• When Engineering Outruns Intelligence: Rethinking Instruction-Guided Navigation
• Trainable Dynamic Mask Sparse Attention
• Communicating Plans, Not Percepts: Scalable Multi-Agent Coordination with Embodied World Models
• The Geometry of Cortical Computation: Manifold Disentanglement and Predictive Dynamics in VCNet
• Sculptor: Empowering LLMs with Cognitive Agency via Active Context Management
• PakBBQ: A Culturally Adapted Bias Benchmark for QA
• BLADE: Block-Sparse Attention Meets Step Distillation for Efficient Video Generation
• Flow Matching for Efficient and Scalable Data Assimilation
• ECHO: Frequency-aware Hierarchical Encoding for Variable-length Signals
• Transduction is All You Need for Structured Data Workflows
• Jet-Nemotron: Efficient Language Model with Post Neural Architecture Search
• Evaluating the Effectiveness of Transformer Layers in Wav2Vec 2.0, XLS-R, and Whisper for Speaker Identification Tasks
• SQL-of-Thought: Multi-agentic Text-to-SQL with Guided Error Correction
• Can General-Purpose Omnimodels Compete with Specialists? A Case Study in Medical Image Segmentation
• Constrained Decoding for Robotics Foundation Models
• MAUSAM: An Observations-focused assessment of Global AI Weather Prediction Models During the South Asian Monsoon
• Do LLMs Adhere to Label Definitions? Examining Their Receptivity to External Label Definitions
• Diffusion Generative Models Meet Compressed Sensing, with Applications to Imaging and Finance
• The Physical Basis of Prediction: World Model Formation in Neural Organoids via an LLM-Generated Curriculum
• Hierarchical Task Environments as the Next Frontier for Embodied World Models in Robot Soccer
• Code2MCP: Transforming Code Repositories into MCP Services
• BranchGRPO: Stable and Efficient GRPO with Structured Branching in Diffusion Models
• COMPACT: Common-token Optimized Model Pruning Across Channels and Tokens
• HealthSLM-Bench: Benchmarking Small Language Models for Mobile and Wearable Healthcare Monitoring
• Do Natural Language Descriptions of Model Activations Convey Privileged Information?
• Imagined Autocurricula
• TreeIRL: Safe Urban Driving with Tree Search and Inverse Reinforcement Learning
• Fast and Fluent Diffusion Language Models via Convolutional Decoding and Rejective Fine-tuning
• Multi-Scenario Highway Lane-Change Intention Prediction: A Physics-Informed AI Framework for Three-Class Classification
• Diversity Boosts AI-Generated Text Detection
• Diffusion-Based Impedance Learning for Contact-Rich Manipulation Tasks
• Experience Deploying Containerized GenAI Services at an HPC Center
• Combinatorial Creativity: A New Frontier in Generalization Abilities
• Can Less Precise Be More Reliable? A Systematic Evaluation of Quantization's Impact on CLIP Beyond Accuracy
• XL-Suite: Cross-Lingual Synthetic Training and Evaluation Data for Open-Ended Generation
• SentenceKV: Efficient LLM Inference via Sentence-Level Semantic KV Caching
• A Practical Synthesis of Detecting AI-Generated Textual, Visual, and Audio Content
• Are You Getting What You Pay For? Auditing Model Substitution in LLM APIs
• Min-Max Optimisation for Nonconvex-Nonconcave Functions Using a Random Zeroth-Order Extragradient Algorithm
• When Federated Learning Meets Quantum Computing: Survey and Research Opportunities
• Evolution Meets Diffusion: Efficient Neural Architecture Generation
• TRACE Back from the Future: A Probabilistic Reasoning Approach to Controllable Language Generation
• Sobolev norm inconsistency of kernel interpolation
• Cooking Up Creativity: Enhancing LLM Creativity through Structured Recombination
• $\textit{New News}$: System-2 Fine-tuning for Robust Integration of New Knowledge
• ReactDance: Hierarchical Representation for High-Fidelity and Coherent Long-Form Reactive Dance Generation
• References Indeed Matter? Reference-Free Preference Optimization for Conversational Query Reformulation
• OnPrem.LLM: A Privacy-Conscious Document Intelligence Toolkit
• The Counting Power of Transformers
• Critique-Guided Distillation for Efficient and Robust Language Model Reasoning
• Fine-grained Contrastive Learning for ECG-Report Alignment with Waveform Enhancement
• AdaBoN: Adaptive Best-of-N Alignment
• FRABench and UFEval: Unified Fine-grained Evaluation with Task and Aspect Generalization
• Is Active Persona Inference Necessary for Aligning Small Models to Personal Preferences?
• TranSUN: A Preemptive Paradigm to Eradicate Retransformation Bias Intrinsically from Regression Models in Recommender Systems
• Mechanistic Fine-tuning for In-context Learning
• Vid2World: Crafting Video Diffusion Models to Interactive World Models
• Internal Chain-of-Thought: Empirical Evidence for Layer-wise Subtask Scheduling in LLMs
• Language Models Optimized to Fool Detectors Still Have a Distinct Style (And How to Change It)
• ReflAct: World-Grounded Decision Making in LLM Agents via Goal-State Reflection
• Pixels Versus Priors: Controlling Knowledge Priors in Vision-Language Models through Visual Counterfacts
• Flexible MOF Generation with Torsion-Aware Flow Matching
• Boosting Open Set Recognition Performance through Modulated Representation Learning
• PoliCon: Evaluating LLMs on Achieving Diverse Political Consensus Objectives
• HS-STaR: Hierarchical Sampling for Self-Taught Reasoners via Difficulty Estimation and Budget Reallocation
• SpecExtend: A Drop-in Enhancement for Speculative Decoding of Long Sequences
• Convergence of Clipped-SGD for Convex $(L_0,L_1)$-Smooth Optimization with Heavy-Tailed Noise
• Unveiling Impact of Frequency Components on Membership Inference Attacks for Diffusion Models
• Cross-modal RAG: Sub-dimensional Text-to-Image Retrieval-Augmented Generation
• ODE-GS: Latent ODEs for Dynamic Scene Extrapolation with 3D Gaussian Splatting
• Neural-Augmented Kelvinlet for Real-Time Soft Tissue Deformation Modeling
• Flexible and Efficient Drift Detection without Labels
• AdversariaL attacK sAfety aLIgnment(ALKALI): Safeguarding LLMs through GRACE: Geometric Representation-Aware Contrastive Enhancement- Introducing Adversarial Vulnerability Quality Index (AVQI)
• When Is Diversity Rewarded in Cooperative Multi-Agent Learning?
• Constant Bit-size Transformers Are Turing Complete
• Do We Need Large VLMs for Spotting Soccer Actions?
• Prover Agent: An Agent-Based Framework for Formal Mathematical Proofs
• R1-Ranker: Teaching LLM Rankers to Reason
• Breaking Rank Bottlenecks in Knowledge Graph Embeddings
• XTransfer: Modality-Agnostic Few-Shot Model Transfer for Human Sensing at the Edge
• Latent Chain-of-Thought? Decoding the Depth-Recurrent Transformer
• Almost Sure Convergence for the Last Iterate of Stochastic Gradient Descent Schemes
• IML-Spikeformer: Input-aware Multi-Level Spiking Transformer for Speech Processing
• Mitigating Watermark Forgery in Generative Models via Randomized Key Selection
• CLEAR: Calibrated Learning for Epistemic and Aleatoric Risk
• Function Induction and Task Generalization: An Interpretability Study with Off-by-One Addition
• Is Thompson Sampling Susceptible to Algorithmic Collusion?
• SEMF: Supervised Expectation-Maximization Framework for Predicting Intervals
• FusionDTI: Fine-grained Binding Discovery with Token-level Fusion for Drug-Target Interaction
• CommonPower: A Framework for Safe Data-Driven Smart Grid Control
• fVDB: A Deep-Learning Framework for Sparse, Large-Scale, and High-Performance Spatial Intelligence
• Optimal thresholds and algorithms for a model of multi-modal learning in high dimensions
• Sheaf Discovery with Joint Computation Graph Pruning and Flexible Granularity
• Bayesian Autoregressive Online Change-Point Detection with Time-Varying Parameters
• A Voter-Based Stochastic Rejection-Method Framework for Asymptotically Safe Language Model Outputs
• Robot Navigation with Entity-Based Collision Avoidance using Deep Reinforcement Learning
• LLM-3D Print: Large Language Models To Monitor and Control 3D Printing
• Attentive Dilated Convolution for Automatic Sleep Staging using Force-directed Layout
• An Empirical Study on the Computation Budget of Co-Optimization of Robot Design and Control in Simulation
• On the Effect of Instability on Learning Continuous-Time Linear Control Systems
• Disentangling Regional Primitives for Image Generation
• LLMs Are In-Context Bandit Reinforcement Learners
• A quantitative Robbins-Siegmund theorem
• CoT-TL: Low-Resource Temporal Knowledge Representation of Planning Instructions Using Chain-of-Thought Reasoning
• PACER: Physics Informed Uncertainty Aware Climate Emulator
• When Speculation Spills Secrets: Side Channels via Speculative Decoding In LLMs
• UniTraj: Learning a Universal Trajectory Foundation Model from Billion-Scale Worldwide Traces
• CART: Compositional Auto-Regressive Transformer for Image Generation
• Gaussian Process Priors for Boundary Value Problems of Linear Partial Differential Equations
• Break the ID-Language Barrier: An Adaption Framework for LLM-based Sequential Recommendation
• Invariant Measures in Time-Delay Coordinates for Unique Dynamical System Identification
• A learning-based approach to stochastic optimal control under reach-avoid constraint
• Order Matters! An Empirical Study on Large Language Models' Input Order Bias in Software Fault Localization
• Reversed in Time: A Novel Temporal-Emphasized Benchmark for Cross-Modal Video-Text Retrieval
• Training-Free Defense Against Adversarial Attacks in Deep Learning MRI Reconstruction
• Improving the adaptive and continuous learning capabilities of artificial neural networks: Lessons from multi-neuromodulatory dynamics
• Gaussian Universality for Diffusion Models
• MIAFEx: An Attention-based Feature Extraction Method for Medical Image Classification
• Nirvana AI Governance: How AI Policymaking Is Committing Three Old Fallacies
• Non-Expansive Mappings in Two-Time-Scale Stochastic Approximation: Finite-Time Analysis
• A Unified Information-Theoretic Framework for Meta-Learning Generalization
• DeepFRC: An End-to-End Deep Learning Model for Functional Registration and Classification
• Half-order Fine-Tuning for Diffusion Model: A Recursive Likelihood Ratio Optimizer
• An Empirical Analysis of Machine Learning Model and Dataset Documentation, Supply Chain, and Licensing Challenges on Hugging Face
• Group-Adaptive Threshold Optimization for Robust AI-Generated Text Detection
• Noise Sensitivity and Learning Lower Bounds for Hierarchical Functions
• OrderFusion: Encoding Orderbook for End-to-End Probabilistic Intraday Electricity Price Forecasting
• Accelerated Parallel Tempering via Neural Transports
• Mitigating Barren Plateaus in Quantum Neural Networks via an AI-Driven Submartingale-Based Framework
• Optimal and Provable Calibration in High-Dimensional Binary Classification: Angular Calibration and Platt Scaling
• Finite-Sample Analysis of Policy Evaluation for Robust Average Reward Reinforcement Learning
• Conformal prediction of future insurance claims in the regression problem
• UniF$^2$ace: A Unified Fine-grained Face Understanding and Generation Model
• A Survey on Self-supervised Contrastive Learning for Multimodal Text-Image Analysis
• Lightweight Learning for Grant-Free Activity Detection in Cell-Free Massive MIMO Networks
• Safety Mirage: How Spurious Correlations Undermine VLM Safety Fine-Tuning and Can Be Mitigated by Machine Unlearning
• A categorical embedding discontinuity-capturing shallow neural network for anisotropic elliptic interface problems
• Machine Learning - Driven Materials Discovery: Unlocking Next-Generation Functional Materials - A review
• Grasping a Handful: Sequential Multi-Object Dexterous Grasp Generation
• SUV: Scalable Large Language Model Copyright Compliance with Regularized Selective Unlearning
• MoQE: Improve Quantization Model performance via Mixture of Quantization Experts
• DHG-Bench: A Comprehensive Benchmark for Deep Hypergraph Learning
• Contrastive Representations for Temporal Reasoning
• Depth-Breadth Synergy in RLVR: Unlocking LLM Reasoning Gains with Adaptive Exploration
• Hard Examples Are All You Need: Maximizing GRPO Post-Training Under Annotation Budgets
• Speculative Safety-Aware Decoding
• Type-Compliant Adaptation Cascades: Adapting Programmatic LM Workflows to Data
• What Matters in Data for DPO?
• End-to-End On-Device Quantization-Aware Training for LLMs at Inference Cost
• T-MLP: Tailed Multi-Layer Perceptron for Level-of-Detail Signal Representation
• Metis: Training Large Language Models with Advanced Low-Bit Quantization
• GradES: Significantly Faster Training in Transformers with Gradient-Based Early Stopping
• Differentiable Expectation-Maximisation and Applications to Gaussian Mixture Model Optimal Transport
• Co-Evolving Complexity: An Adversarial Framework for Automatic MARL Curricula
• Small Vectors, Big Effects: A Mechanistic Study of RL-Induced Reasoning via Steering Vectors
• Graph Alignment via Dual-Pass Spectral Encoding and Latent Space Communication
• Learning to Route: Per-Sample Adaptive Routing for Multimodal Multitask Prediction
• TDRM: Smooth Reward Models with Temporal Difference for LLM RL and Inference
• Enhancing Generative Auto-bidding with Offline Reward Evaluation and Policy Search
• VQEzy: An Open-Source Dataset for Parameter Initialization in Variational Quantum Eigensolvers
• Joint Memory Frequency and Computing Frequency Scaling for Energy-efficient DNN Inference
• PiERN: Token-Level Routing for Integrating High-Precision Computation and Reasoning
• PipelineRL: Faster On-policy Reinforcement Learning for Long Sequence Generation
• Network inference via process motifs for lagged correlation in linear stochastic processes
• Explicit Second-Order Min-Max Optimization: Practical Algorithms and Complexity Analysis
• Policy Gradient Algorithms for Robust MDPs with Non-Rectangular Uncertainty Sets
• A Double Machine Learning Approach to Combining Experimental and Observational Data
• Few-shot Personalized Saliency Prediction Based on Interpersonal Gaze Patterns
• Information theory for data-driven model reduction in physics and biology
• A Proximal Gradient Method With Probabilistic Multi-Gossip Communications for Decentralized Composite Optimization
• Machines Do See Color: A Guideline to Classify Different Forms of Racist Discourse in Large Corpora
• Off-Policy Evaluation in Markov Decision Processes under Weak Distributional Overlap
• Towards Efficient LLM Grounding for Embodied Multi-Agent Collaboration
• Angles Don't Lie: Unlocking Training-Efficient RL Through the Model's Own Signals
• Comba: Improving Bilinear RNNs with Closed-loop Control
• QKV Projections Require a Fraction of Their Memory
• Interaction Field Matching: Overcoming Limitations of Electrostatic Models
• Curse of Slicing: Why Sliced Mutual Information is a Deceptive Measure of Statistical Dependence
• Bridging the Performance Gap Between Target-Free and Target-Based Reinforcement Learning
• Leveraging Coordinate Momentum in SignSGD and Muon: Memory-Optimized Zero-Order
• Towards Better Generalization via Distributional Input Projection Network
• Reshaping Reasoning in LLMs: A Theoretical Analysis of RL Training Dynamics through Pattern Selection
• TreeRPO: Tree Relative Policy Optimization
• Flow-Attentional Graph Neural Networks
• Can In-Context Reinforcement Learning Recover From Reward Poisoning Attacks?
• InverseScope: Scalable Activation Inversion for Interpreting Large Language Models
• Why Masking Diffusion Works: Condition on the Jump Schedule for Improved Discrete Diffusion
• Towards Robust Real-World Multivariate Time Series Forecasting: A Unified Framework for Dependency, Asynchrony, and Missingness
• InstructPro: Natural Language Guided Ligand-Binding Protein Design
• Foundation Models for Causal Inference via Prior-Data Fitted Networks
• Enhancing Delta Compression in LLMs via SVD-based Quantization Error Minimization
• Meta Pruning via Graph Metanetworks : A Universal Meta Learning Framework for Network Pruning
• Unlearning Isn't Invisible: Detecting Unlearning Traces in LLMs from Model Outputs
• Muon Optimizes Under Spectral Norm Constraints
• Reward-Agnostic Prompt Optimization for Text-to-Image Diffusion Models
• Adaptive Sample Scheduling for Direct Preference Optimization
• Origins of Creativity in Attention-Based Diffusion Models
• Aha Moment Revisited: Are VLMs Truly Capable of Self Verification in Inference-time Scaling?
• Mitigating Semantic Collapse in Generative Personalization with Test-Time Embedding Adjustment
• Data Uniformity Improves Training Efficiency and More, with a Convergence Framework Beyond the NTK Regime
• Theoretical Modeling of LLM Self-Improvement Training Dynamics Through Solver-Verifier Gap
• Cooperative Sheaf Neural Networks
• Learning to Segment for Vehicle Routing Problems
• JAX-MPM: A Learning-Augmented Differentiable Meshfree Framework for GPU-Accelerated Lagrangian Simulation and Geophysical Inverse Modeling
• Model Collapse Is Not a Bug but a Feature in Machine Unlearning for LLMs
• Discrete Diffusion Trajectory Alignment via Stepwise Decomposition
• Robust Deep Network Learning of Nonlinear Regression Tasks by Parametric Leaky Exponential Linear Units (LELUs) and a Diffusion Metric
• One Token to Fool LLM-as-a-Judge
• Warm Starts Accelerate Conditional Diffusion
• FusionFactory: Fusing LLM Capabilities with Multi-LLM Log Data
• A Graph-in-Graph Learning Framework for Drug-Target Interaction Prediction
• Vidar: Embodied Video Diffusion Model for Generalist Manipulation
• Probabilistic Soundness Guarantees in LLM Reasoning Chains
• Learning to summarize user information for personalized reinforcement learning from human feedback
• GRID: Scalable Task-Agnostic Prompt-Based Continual Learning for Language Models
• Omni-Thinker: Scaling Multi-Task RL in LLMs with Hybrid Reward and Task Scheduling
• U-Cast: Learning Hierarchical Structures for High-Dimensional Time Series Forecasting
• Enhancing Stability of Physics-Informed Neural Network Training Through Saddle-Point Reformulation
• GLANCE: Graph Logic Attention Network with Cluster Enhancement for Heterophilous Graph Representation Learning
• Moving Out: Physically-grounded Human-AI Collaboration
• A Markov Categorical Framework for Language Modeling
• Can Language Models Discover Scaling Laws?
• Merging Memory and Space: A State Space Neural Operator
• Signals, Concepts, and Laws: Toward Universal, Explainable Time-Series Forecasting
• Beyond Prompt-Induced Lies: Investigating LLM Deception on Benign Prompts
• Runtime Adaptive Pruning for LLM Inference
• On the Design of KL-Regularized Policy Gradient Algorithms for LLM Reasoning
• Trinity-RFT: A General-Purpose and Unified Framework for Reinforcement Fine-Tuning of Large Language Models
• What Do You Need for Diverse Trajectory Composition in Diffusion Planning?
• Reward Model Overoptimisation in Iterated RLHF
• Improved Sample Complexity For Diffusion Model Training Without Empirical Risk Minimizer Access
• ThanoRA: Task Heterogeneity-Aware Multi-Task Low-Rank Adaptation
• LoTA-QAF: Lossless Ternary Adaptation for Quantization-Aware Fine-Tuning
• Diagnosing and Mitigating Modality Interference in Multimodal Large Language Models
• Logic Gate Neural Networks are Good for Verification
• ePC: Overcoming Exponential Signal Decay in Deep Predictive Coding Networks
• Variational Deep Learning via Implicit Regularization
• PDFBench: A Benchmark for De novo Protein Design from Function
• Stochastic Primal-Dual Double Block-Coordinate for Two-way Partial AUC Maximization
• Equivariant Spherical Transformer for Efficient Molecular Modeling
• Efficient AllReduce with Stragglers
• Continuous Chain of Thought Enables Parallel Exploration and Reasoning
• Vision Language Models are Biased
• On the Emergence of Weak-to-Strong Generalization: A Bias-Variance Perspective
• ReCalKV: Low-Rank KV Cache Compression via Head Reordering and Offline Calibration
• Weight-Space Linear Recurrent Neural Networks
• Chunking the Critic: A Transformer-based Soft Actor-Critic with N-Step Returns
• What Makes a Reward Model a Good Teacher? An Optimization Perspective
• On The Sample Complexity Bounds In Bilevel Reinforcement Learning
• Finite-Time Bounds for Two-Time-Scale Stochastic Approximation with Arbitrary Norm Contractions and Markovian Noise
• Reasoning to Learn from Latent Thoughts
• AdaRank: Adaptive Rank Pruning for Enhanced Model Merging
• Pairwise Optimal Transports for Training All-to-All Flow-Based Condition Transfer Model
• Optuna vs Code Llama: Are LLMs a New Paradigm for Hyperparameter Tuning?
• Efficient Generative Model Training via Embedded Representation Warmup
• A Model Zoo on Phase Transitions in Neural Networks
• A Unified MDL-based Binning and Tensor Factorization Framework for PDF Estimation
• Sharpness-Aware Minimization with Z-Score Gradient Filtering
• Localized Diffusion Models
• Generating Full-field Evolution of Physical Dynamics from Irregular Sparse Observations
• RainPro-8: An Efficient Deep Learning Model to Estimate Rainfall Probabilities Over 8 Hours
• Fine-tuning Diffusion Policies with Backpropagation Through Diffusion Timesteps
• The Final Layer Holds the Key: A Unified and Efficient GNN Calibration Framework
• Visual Planning: Let's Think Only with Images
• Stepwise Guided Policy Optimization: Coloring your Incorrect Reasoning in GRPO
• Continuous Optimization for Feature Selection with Permutation-Invariant Embedding and Policy-Guided Search
• MINGLE: Mixture of Null-Space Gated Low-Rank Experts for Test-Time Continual Model Merging
• AltLoRA: Towards Better Gradient Approximation in Low-Rank Adaptation with Alternating Projections
• Hamiltonian Neural PDE Solvers through Functional Approximation
• Causes and Consequences of Representational Similarity in Machine Learning Models
• VAMO: Efficient Zeroth-Order Variance Reduction for SGD with Faster Convergence
• Learning with Local Search MCMC Layers
• PiFlow: Principle-aware Scientific Discovery with Multi-Agent Collaboration
• Scaling Diffusion Transformers Efficiently via $\mu$P
• Certified Neural Approximations of Nonlinear Dynamics
• Towards Identifiability of Interventional Stochastic Differential Equations
• Not All Models Suit Expert Offloading: On Local Routing Consistency of Mixture-of-Expert Models
• Scalable Graph Generative Modeling via Substructure Sequences
• AdaSTaR: Adaptive Data Sampling for Training Self-Taught Reasoners
• ICYM2I: The illusion of multimodal informativeness under missingness
• Generalized Tangent Kernel: A Unified Geometric Foundation for Natural Gradient and Standard Gradient
• TabText: Language-Based Representations of Tabular Health Data for Predictive Modelling
• Beyond Losses Reweighting: Empowering Multi-Task Learning via the Generalization Perspective
• AQuaMaM: An Autoregressive, Quaternion Manifold Model for Rapidly Estimating Complex SO(3) Distributions
• Symbolic Imitation Learning: From Black-Box to Explainable Driving Policies
• Double Machine Learning Based Structure Identification from Temporal Data
• EUGENE: Explainable Structure-aware Graph Edit Distance Estimation with Generalized Edit Costs
• The Clever Hans Mirage: A Comprehensive Survey on Spurious Correlations in Machine Learning
• Federated Learning Resilient to Byzantine Attacks and Data Heterogeneity
• Data Imputation by Pursuing Better Classification: A Supervised Kernel-Based Method
• PLEIADES: Building Temporal Kernels with Orthogonal Polynomials
• A Comprehensive Graph Pooling Benchmark: Effectiveness, Robustness and Generalizability
• Differential Encoding for Improved Representation Learning over Graphs
• Deep Time Series Models: A Comprehensive Survey and Benchmark
• Can DPO Learn Diverse Human Values? A Theoretical Scaling Law
• Understanding Transformer Architecture through Continuous Dynamics: A Partial Differential Equation Perspective
• Efficient Federated Learning against Byzantine Attacks and Data Heterogeneity via Aggregating Normalized Gradients
• A GREAT Architecture for Edge-Based Graph Problems Like TSP
• Sparse Covariance Neural Networks
• DeepONet for Solving Nonlinear Partial Differential Equations with Physics-Informed Training
• Extracting Moore Machines from Transformers using Queries and Counterexamples
• Functional-level Uncertainty Quantification for Calibrated Fine-tuning on LLMs
• Deeper Insights into Deep Graph Convolutional Networks: Stability and Generalization
• NextLocLLM: Location Semantics Modeling and Coordinate-Based Next Location Prediction with LLMs
• Gradient-Free Training of Quantized Neural Networks
• Similarity-Dissimilarity Loss for Multi-label Supervised Contrastive Learning
• A Predictive Approach To Enhance Time-Series Forecasting
• Benchmarking Computational Methods for Emerging Drug-Drug Interaction Prediction
• Self-Normalized Resets for Plasticity in Continual Learning
• Haar-Laplacian for directed graphs
• Sometimes I am a Tree: Data Drives Unstable Hierarchical Generalization
• Off-Policy Maximum Entropy RL with Future State and Action Visitation Measures
• Euclidean Fast Attention - Machine Learning Global Atomic Representations at Linear Cost
• Learning Randomized Reductions
• Toward Model-centric Heterogeneous Federated Graph Learning: A Knowledge-driven Approach
• Norm-Bounded Low-Rank Adaptation
• Principal Components for Neural Network Initialization
• Federated Sketching LoRA: A Flexible Framework for Heterogeneous Collaborative Fine-Tuning of LLMs
• Vintix: Action Model via In-Context Reinforcement Learning
• DAL: A Practical Prior-Free Black-Box Framework for Non-Stationary Bandits
• A Statistical Learning Perspective on Semi-dual Adversarial Neural Optimal Transport Solvers
• InfoBridge: Mutual Information estimation via Bridge Matching
• LEAD: Large Foundation Model for EEG-Based Alzheimer's Disease Detection
• Progressive Binarization with Semi-Structured Pruning for LLMs
• Towards Large-Scale In-Context Reinforcement Learning by Meta-Training in Randomized Worlds
• Pre-training Epidemic Time Series Forecasters with Compartmental Prototypes
• vCache: Verified Semantic Prompt Caching
• Functional Complexity-adaptive Temporal Tensor Decomposition
• Recurrent Memory for Online Interdomain Gaussian Processes
• The Accuracy Cost of Weakness: A Theoretical Analysis of Fixed-Segment Weak Labeling for Events in Time
• Comprehensive Review of Neural Differential Equations for Time Series Analysis
• Learning to Explain Air Traffic Situation
• Collaborative Deterministic-Probabilistic Forecasting for Diverse Spatiotemporal Systems
• SURGE: On the Potential of Large Language Models as General-Purpose Surrogate Code Executors
• TGT: A Temporal Gating Transformer for Smartphone App Usage Prediction
• Joint Value Estimation and Bidding in Repeated First-Price Auctions
• Meta-Learning to Explore via Memory Density Feedback
• Neuroplasticity-inspired dynamic ANNs for multi-task demand forecasting
• CMT: Mid-Training for Efficient Learning of Consistency, Mean Flow, and Flow Map Models
• Training Agents Inside of Scalable World Models
• Quantitative convergence of trained single layer neural networks to Gaussian processes
• Bandits roaming Hilbert space
• Prompting Robot Teams with Natural Language
• Inducing Dyslexia in Vision Language Models
• Algorithms and data structures for automatic precision estimation of neural networks
• Hype or not? Formalizing Automatic Promotional Language Detection in Biomedical Research
• InfLLM-V2: Dense-Sparse Switchable Attention for Seamless Short-to-Long Adaptation
• Reference-Free Rating of LLM Responses via Latent Information
• Stabilizing Humanoid Robot Trajectory Generation via Physics-Informed Learning and Control-Informed Steering
• MAD: Manifold Attracted Diffusion
• Bundle Network: a Machine Learning-Based Bundle Method
• ProxyAttn: Guided Sparse Attention via Representative Heads
• Spatial-Functional awareness Transformer-based graph archetype contrastive learning for Decoding Visual Neural Representations from EEG
• Sparse Autoencoders Make Audio Foundation Models more Explainable
• Fidelity-Aware Data Composition for Robust Robot Generalization
• TACO-Net: Topological Signatures Triumph in 3D Object Classification
• A Greedy PDE Router for Blending Neural Operators and Classical Methods
• Efficient Sketching and Nearest Neighbor Search Algorithms for Sparse Vector Sets
• Of-SemWat: High-payload text embedding for semantic watermarking of AI-generated images with arbitrary size
• Pushing LLMs to Their Logical Reasoning Bound: The Role of Data Reasoning Intensity
• Environment-Aware Satellite Image Generation with Diffusion Models
• VAGUEGAN: Stealthy Poisoning and Backdoor Attacks on Image Generative Pipelines
• Improved Stochastic Optimization of LogSumExp
• Unmute the Patch Tokens: Rethinking Probing in Multi-Label Audio Classification
• When Scores Learn Geometry: Rate Separations under the Manifold Hypothesis
• Inductive Bias and Spectral Properties of Single-Head Attention in High Dimensions
• From Code to Action: Hierarchical Learning of Diffusion-VLM Policies
• A Spectral-Grassmann Wasserstein metric for operator representations of dynamical systems
• Graph Theory Meets Federated Learning over Satellite Constellations: Spanning Aggregations, Network Formation, and Performance Optimization
• Scalable GANs with Transformers
• MSG: Multi-Stream Generative Policies for Sample-Efficient Robotic Manipulation
• Embedded Deep Learning for Bio-hybrid Plant Sensors to Detect Increased Heat and Ozone Levels
• LVT: Large-Scale Scene Reconstruction via Local View Transformers
• CLASP: Adaptive Spectral Clustering for Unsupervised Per-Image Segmentation
• VT-FSL: Bridging Vision and Text with LLMs for Few-Shot Learning
• Ultra-Fast Language Generation via Discrete Diffusion Divergence Instruct
• Hyperdimensional Probe: Decoding LLM Representations via Vector Symbolic Architectures
• Symmetry-Aware Bayesian Optimization via Max Kernels
• Cogito, Ergo Ludo: An Agent that Learns to Play by Reasoning and Planning
• Optimizing Privacy-Preserving Primitives to Support LLM-Scale Applications
• Scaling Generalist Data-Analytic Agents
• Benchmarking ECG Foundational Models: A Reality Check Across Clinical Tasks
• Curriculum Imitation Learning of Distributed Multi-Robot Policies
• On Spectral Learning for Odeco Tensors: Perturbation, Initialization, and Algorithms
• Score Distillation of Flow Matching Models
• The Era of Real-World Human Interaction: RL from User Conversations
• Paired by the Teacher: Turning Unpaired Data into High-Fidelity Pairs for Low-Resource Text Generation
• Fast Feature Field ($\text{F}^3$): A Predictive Representation of Events
• Pretraining Large Language Models with NVFP4
• Context-Driven Performance Modeling for Causal Inference Operators on Neural Processing Units
• Personalized Vision via Visual In-Context Learning
• GHOST: Hallucination-Inducing Image Generation for Multimodal LLMs
• Last iterate convergence in no-regret learning: constrained min-max optimization for convex-concave landscapes
• CRAUM-Net: Contextual Recursive Attention with Uncertainty Modeling for Salient Object Detection
• DFG-PCN: Point Cloud Completion with Degree-Flexible Point Graph
• StrucADT: Generating Structure-controlled 3D Point Clouds with Adjacency Diffusion Transformer
• Do LLMs Understand Romanian Driving Laws? A Study on Multimodal and Fine-Tuned Question Answering
• Diff-3DCap: Shape Captioning with Diffusion Models
• LUQ: Layerwise Ultra-Low Bit Quantization for Multimodal Large Language Models
• VioPTT: Violin Technique-Aware Transcription from Synthetic Data Augmentation
• Knowledge-Level Consistency Reinforcement Learning: Dual-Fact Alignment for Long-Form Factuality
• Sequence Pathfinder for Multi-Agent Pickup and Delivery in the Warehouse
• Define latent spaces by example: optimisation over the outputs of generative models
• Influence-Guided Concolic Testing of Transformer Robustness
• A Multi-Camera Vision-Based Approach for Fine-Grained Assembly Quality Control
• Assessing Visual Privacy Risks in Multimodal AI: A Novel Taxonomy-Grounded Evaluation of Vision-Language Models
• Taught Well Learned Ill: Towards Distillation-conditional Backdoor Attack
• Learning-Based Testing for Deep Learning: Enhancing Model Robustness with Adversarial Input Prioritization
• Equation-Free Coarse Control of Distributed Parameter Systems via Local Neural Operators
• Toward Preference-aligned Large Language Models via Residual-based Model Steering
• TREAT-Net: Tabular-Referenced Echocardiography Analysis for Acute Coronary Syndrome Treatment Prediction
• Sequential Diffusion Language Models
• The Role of Logic and Automata in Understanding Transformers
• Singleton-Optimized Conformal Prediction
• GEAR: A General Evaluation Framework for Abductive Reasoning
• SpeedCP: Fast Kernel-based Conditional Conformal Prediction
• Fathom-DeepResearch: Unlocking Long Horizon Information Retrieval and Synthesis for SLMs
• Ancestry Tree Clustering for Particle Filter Diversity Maintenance
• ASTROCO: Self-Supervised Conformer-Style Transformers for Light-Curve Embeddings
• EYE-DEX: Eye Disease Detection and EXplanation System
• Your thoughts tell who you are: Characterize the reasoning patterns of LRMs
• Accelerating Cerebral Diagnostics with BrainFusion: A Comprehensive MRI Tumor Framework
• STRAPSim: A Portfolio Similarity Metric for ETF Alignment and Portfolio Trades
• Memory Transfer Planning: LLM-driven Context-Aware Code Adaptation for Robot Manipulation
• Retrieval-augmented GUI Agents with Generative Guidelines
• AceSearcher: Bootstrapping Reasoning and Search for LLMs via Reinforced Self-Play
• BeyondBench: Benchmark-Free Evaluation of Reasoning in Language Models
• ViReSkill: Vision-Grounded Replanning with Skill Memory for LLM-Based Planning in Lifelong Robot Learning
• Uni-NTFM: A Unified Foundation Model for EEG Signal Representation Learning
• Non-Invasive Detection of PROState Cancer with Novel Time-Dependent Diffusion MRI and AI-Enhanced Quantitative Radiological Interpretation: PROS-TD-AI
• SpecExit: Accelerating Large Reasoning Model via Speculative Exit
• Interactive Program Synthesis for Modeling Collaborative Physical Activities from Narrated Demonstrations
• Extracting the Structure of Press Releases for Predicting Earnings Announcement Returns
• Understanding Cognitive States from Head & Hand Motion Data
• VeriLLM: A Lightweight Framework for Publicly Verifiable Decentralized Inference
• Risk-Sensitive RL for Alleviating Exploration Dilemmas in Large Language Models
• LAMP-PRo: Label-aware Attention for Multi-label Prediction of DNA- and RNA-binding Proteins using Protein Language Models
• Graph-Based Learning of Free Surface Dynamics in Generalized Newtonian Fluids using Smoothed Particle Hydrodynamics
• Skeleton-based Robust Registration Framework for Corrupted 3D Point Clouds
• SCI-Verifier: Scientific Verifier with Thinking
• ActiveCQ: Active Estimation of Causal Quantities
• PEARL: Performance-Enhanced Aggregated Representation Learning
• Inferring Cosmological Parameters with Evidential Physics-Informed Neural Networks
• Hyperspherical Latents Improve Continuous-Token Autoregressive Generation
• DRIFT: Divergent Response in Filtered Transformations for Robust Adversarial Defense
• Prediction-Powered Communication with Distortion Guarantees
• From Sound to Setting: AI-Based Equalizer Parameter Prediction for Piano Tone Replication
• FuncPoison: Poisoning Function Library to Hijack Multi-agent Autonomous Driving Systems
• Multi-Item-Query Attention for Stable Sequential Recommendation
• Contrastive Learning for Correlating Network Incidents
• Euclid's Gift: Enhancing Spatial Perception and Reasoning in Vision-Language Models via Geometric Surrogate Tasks
• Sanitize Your Responses: Mitigating Privacy Leakage in Large Language Models
• Overcoming Over-Fitting in Constraint Acquisition via Query-Driven Interactive Refinement
• Preference-Based Dynamic Ranking Structure Recognition
• EKF-Based Fusion of Wi-Fi/LiDAR/IMU for Indoor Localization and Navigation
• Impact of Environmental Factors on LoRa 2.4 GHz Time of Flight Ranging Outdoors
• Statistical Inference for Gradient Boosting Regression
• Conditional Risk Minimization with Side Information: A Tractable, Universal Optimal Transport Framework
• MathBode: Frequency-Domain Fingerprints of LLM Mathematical Reasoning
• Tree Reward-Aligned Search for TReASURe in Masked Diffusion Language Models
• AI-Enhanced Distributed Channel Access for Collision Avoidance in Future Wi-Fi 8
• Grouped Satisficing Paths in Pure Strategy Games: a Topological Perspective
• Understanding and Enhancing the Planning Capability of Language Models via Multi-Token Prediction
• UltraUNet: Real-Time Ultrasound Tongue Segmentation for Diverse Linguistic and Imaging Conditions
• A Generative Model for Controllable Feature Heterophily in Graphs
• Explicit modelling of subject dependency in BCI decoding
• Learning Regional Monsoon Patterns with a Multimodal Attention U-Net
• Scaling Policy Compliance Assessment in Language Models with Policy Reasoning Traces
• Multifractal features of multimodal cardiac signals: Nonlinear dynamics of exercise recovery
• Space Robotics Bench: Robot Learning Beyond Earth
• Targeted perturbations reveal brain-like local coding axes in robustified, but not standard, ANN-based brain models
• PARROT: A Benchmark for Evaluating LLMs in Cross-System SQL Translation
• CrediBench: Building Web-Scale Network Datasets for Information Integrity
• AI-Assisted Music Production: A User Study on Text-to-Music Models
• Alignment through Meta-Weighted Online Sampling: Bridging the Gap between Data Generation and Preference Optimization
• An Accelerated Newton-GMRES Method for Multilinear PageRank
• Train Once, Answer All: Many Pretraining Experiments for the Cost of One
• Flow Matching for Robust Simulation-Based Inference under Model Misspecification
• Optimizing the Network Topology of a Linear Reservoir Computer
• Comparison of Scoring Rationales Between Large Language Models and Human Raters
• Democratizing AI scientists using ToolUniverse
• New Insights and Algorithms for Optimal Diagonal Preconditioning
• S$^3$F-Net: A Multi-Modal Approach to Medical Image Classification via Spatial-Spectral Summarizer Fusion Network
• AudioFuse: Unified Spectral-Temporal Learning via a Hybrid ViT-1D CNN Architecture for Robust Phonocardiogram Classification
• 3DPCNet: Pose Canonicalization for Robust Viewpoint-Invariant 3D Kinematic Analysis from Monocular RGB cameras
• Multi-Modal Manipulation via Multi-Modal Policy Consensus
• Dynamic Trust Calibration Using Contextual Bandits
• Multi-modal Data Spectrum: Multi-modal Datasets are Multi-dimensional
• Network-Optimised Spiking Neural Network for Event-Driven Networking
• On the Shelf Life of Fine-Tuned LLM Judges: Future Proofing, Backward Compatibility, and Question Generalization
• End-to-End Deep Learning for Predicting Metric Space-Valued Outputs
• RAVEN: Resilient Aerial Navigation via Open-Set Semantic Memory and Behavior Adaptation
• Node Classification via Simplicial Interaction with Augmented Maximal Clique Selection
• BridgeDrive: Diffusion Bridge Policy for Closed-Loop Trajectory Planning in Autonomous Driving
• Large Language Models and Futures Price Factors in China
• Spatially Parallel All-optical Neural Networks
• Communication-aware Wide-Area Damping Control using Risk-Constrained Reinforcement Learning
• RIV: Recursive Introspection Mask Diffusion Vision Language Model
• How LLMs Learn to Reason: A Complex Network Perspective
• LightFair: Towards an Efficient Alternative for Fair T2I Diffusion via Debiasing Pre-trained Text Encoders
• Focusing on What Matters: Object-Agent-centric Tokenization for Vision Language Action models
• Confidence Aware SSD Ensemble with Weighted Boxes Fusion for Weapon Detection
• FMC-DETR: Frequency-Decoupled Multi-Domain Coordination for Aerial-View Object Detection
• Risk Profiling and Modulation for LLMs
• Sparse Deep Additive Model with Interactions: Enhancing Interpretability and Predictability
• FedBit: Accelerating Privacy-Preserving Federated Learning via Bit-Interleaved Packing and Cross-Layer Co-Design
• How to Make Large Language Models Generate 100% Valid Molecules?
• Physics-Informed Inductive Biases for Voltage Prediction in Distribution Grids
• GLASS Flows: Transition Sampling for Alignment of Flow and Diffusion Models
• TR2-D2: Tree Search Guided Trajectory-Aware Fine-Tuning for Discrete Diffusion
• XQC: Well-conditioned Optimization Accelerates Deep Reinforcement Learning
• SIRI: Scaling Iterative Reinforcement Learning with Interleaved Compression
• Exploring Large Language Models for Translating Romanian Computational Problems into English
• BacPrep: An Experimental Platform for Evaluating LLM-Based Bacalaureat Assessment
• Leveraging Generative AI for Enhancing Automated Assessment in Programming Education Contests
• A Culturally-Rich Romanian NLP Dataset from "Who Wants to Be a Millionaire?" Videos
• VOCABTRIM: Vocabulary Pruning for Efficient Speculative Decoding in LLMs
• MateInfoUB: A Real-World Benchmark for Testing LLMs in Competitive, Multilingual, and Multimodal Educational Tasks
• GRILE: A Benchmark for Grammar Reasoning and Explanation in Romanian LLMs
• Stable and Interpretable Jet Physics with IRC-Safe Equivariant Feature Extraction
• A Comprehensive Analysis of Churn Prediction in Telecommunications Using Machine Learning
• Forecasting West Nile virus with deep graph encoders
• A Comparison of Surrogate Constitutive Models for Viscoplastic Creep Simulation of HT-9 Steel
• Semantic-Aware Edge Intelligence for UAV Handover in 6G Networks
• PISA: An AI Pipeline for Interpretable-by-design Survival Analysis Providing Multiple Complexity-Accuracy Trade-off Models
• Profit over Proxies: A Scalable Bayesian Decision Framework for Optimizing Multi-Variant Online Experiments
• Learning Hyperspectral Images with Curated Text Prompts for Efficient Multimodal Alignment
• Enhancing Cluster Scheduling in HPC: A Continuous Transfer Learning for Real-Time Optimization
• Metadata-Guided Adaptable Frequency Scaling across Heterogeneous Applications and Devices
• GZSL-MoE: Apprentissage G{\'e}n{\'e}ralis{\'e} Z{\'e}ro-Shot bas{\'e} sur le M{\'e}lange d'Experts pour la Segmentation S{\'e}mantique de Nuages de Points 3DAppliqu{\'e} {\`a} un Jeu de Donn{\'e}es d'Environnement de Collaboration Humain-Robot
• IBiT: Utilizing Inductive Biases to Create a More Data Efficient Attention Mechanism
• LayoutAgent: A Vision-Language Agent Guided Compositional Diffusion for Spatial Layout Planning
• A Data-Driven Framework for Digital Transformation in Smart Cities: Integrating AI, Dashboards, and IoT Readiness
• Consistency Models as Plug-and-Play Priors for Inverse Problems
• Enabling Approximate Joint Sampling in Diffusion LMs
• Painless Activation Steering: An Automated, Lightweight Approach for Post-Training Large Language Models
• Generalization Analysis for Classification on Korobov Space
• Variance-Bounded Evaluation without Ground Truth: VB-Score
• Concept activation vectors: a unifying view and adversarial attacks
• Identifying Memory Effects in Epidemics via a Fractional SEIRD Model and Physics-Informed Neural Networks
• UESA-Net: U-Shaped Embedded Multidirectional Shrinkage Attention Network for Ultrasound Nodule Segmentation
• A theoretical guarantee for SyncRank
• Differentially Private Two-Stage Gradient Descent for Instrumental Variable Regression
• What Do They Fix? LLM-Aided Categorization of Security Patches for Critical Memory Bugs
• Hilbert: Recursively Building Formal Proofs with Informal Reasoning
• Efficient Fine-Grained GPU Performance Modeling for Distributed Deep Learning of LLM
• Text-Independent Speaker Identification Using Audio Looping With Margin Based Loss Functions
• Learning Temporal Saliency for Time Series Forecasting with Cross-Scale Attention
• Learning to Detect Relevant Contexts and Knowledge for Response Selection in Retrieval-based Dialogue Systems
• Parameterized Hardness of Zonotope Containment and Neural Network Verification
• Patient-specific Biomolecular Instruction Tuning
• Ringleader ASGD: The First Asynchronous SGD with Optimal Time Complexity under Data Heterogeneity
• HEART: Emotionally-driven test-time scaling of Language Models
• Mixtures Closest to a Given Measure: A Semidefinite Programming Approach
• Convolutional Set Transformer
• A benchmark for vericoding: formally verified program synthesis
• TY-RIST: Tactical YOLO Tricks for Real-time Infrared Small Target Detection
• Label-Guided Imputation via Forest-Based Proximities for Improved Time Series Classification
• Soft-Di[M]O: Improving One-Step Discrete Image Generation with Soft Embeddings
• Localized Uncertainty Quantification in Random Forests via Proximities
• What Matters More For In-Context Learning under Matched Compute Budgets: Pretraining on Natural Text or Incorporating Targeted Synthetic Examples?
• Robot Learning from Any Images
• ADAM: A Diverse Archive of Mankind for Evaluating and Enhancing LLMs in Biographical Reasoning
• Unsupervised Conformal Inference: Bootstrapping and Alignment to Control LLM Uncertainty
• Activation Matching for Explanation Generation
• Deep Reinforcement Learning in Action: Real-Time Control of Vortex-Induced Vibrations
• Emergent World Representations in OpenVLA
• Learning to Solve Optimization Problems Constrained with Partial Differential Equations
• SAIP: A Plug-and-Play Scale-adaptive Module in Diffusion-based Inverse Problems
• CURA: Size Isnt All You Need - A Compact Universal Architecture for On-Device Intelligence
• Evaluating classification performance across operating contexts: A comparison of decision curve analysis and cost curves
• OrthAlign: Orthogonal Subspace Decomposition for Non-Interfering Multi-Objective Alignment
• Learning Hamiltonian Dynamics at Scale: A Differential-Geometric Approach
• Identity Bridge: Enabling Implicit Reasoning via Shared Latent Memory
• HyperHELM: Hyperbolic Hierarchy Encoding for mRNA Language Modeling
• T-POP: Test-Time Personalization with Online Preference Feedback
• FedPOB: Sample-Efficient Federated Prompt Optimization via Bandits
• Circuit-Aware Reward Training: A Mechanistic Framework for Longtail Robustness in RLHF
• Discrete Variational Autoencoding via Policy Search
• Q-Net: Transferable Queue Length Estimation via Kalman-based Neural Networks
• Beyond Softmax: A Natural Parameterization for Categorical Random Variables
• Who invented deep residual learning?
• A TRIANGLE Enables Multimodal Alignment Beyond Cosine Similarity
• Robust Policy Expansion for Offline-to-Online RL under Diverse Data Corruption
• In-Context Learning of Temporal Point Processes with Foundation Inference Models
• Neural Message-Passing on Attention Graphs for Hallucination Detection
• MarS-FM: Generative Modeling of Molecular Dynamics via Markov State Models
• Quantifying Generalisation in Imitation Learning
• Assessing the risk of future Dunkelflaute events for Germany using generative deep learning
• Fidel-TS: A High-Fidelity Benchmark for Multimodal Time Series Forecasting
• DSAT-HD: Dual-Stream Adaptive Transformer with Hybrid Decomposition for Multivariate Time Series Forecasting
• Physics-informed learning under mixing: How physical knowledge speeds up learning
• DyMoDreamer: World Modeling with Dynamic Modulation
• Putnam-like dataset summary: LLMs as mathematical competition contestants
• Cell2Text: Multimodal LLM for Generating Single-Cell Descriptions from RNA-Seq Data
• Beyond the Hook: Predicting Billboard Hot 100 Chart Inclusion with Machine Learning from Streaming, Audio Signals, and Perceptual Features
• DRIFT-Net: A Spectral--Coupled Neural Operator for PDEs Learning
• Uncertainty-Guided Expert-AI Collaboration for Efficient Soil Horizon Annotation
• Scaling Laws and Spectra of Shallow Neural Networks in the Feature Learning Regime
• Adaptive Canonicalization with Application to Invariant Anisotropic Geometric Networks
• Towards Understanding the Shape of Representations in Protein Language Models
• When Greedy Wins: Emergent Exploitation Bias in Meta-Bandit LLM Training
• Is Sequence Information All You Need for Bayesian Optimization of Antibodies?
• OAT-FM: Optimal Acceleration Transport for Improved Flow Matching
• Learning Distinguishable Representations in Deep Q-Networks for Linear Transfer
• Intra-request branch orchestration for efficient LLM reasoning
• Overlap-Adaptive Regularization for Conditional Average Treatment Effect Estimation
• Double Descent as a Lens for Sample Efficiency in Autoregressive vs. Discrete Diffusion Models
• Random Policy Valuation is Enough for LLM Reasoning with Verifiable Rewards
• Sampling Complexity of TD and PPO in RKHS
• Score-based Membership Inference on Diffusion Models
• Uncertainty-Aware Deep Learning for Wildfire Danger Forecasting
• MARCOS: Deep Thinking by Markov Chain of Continuous Thoughts
• Bayesian Surrogates for Risk-Aware Pre-Assessment of Aging Bridge Portfolios
• A multiscale analysis of mean-field transformers in the moderate interaction regime
• Efficient Hyperparameter Tuning via Trajectory Invariance Principle
• Advantage Weighted Matching: Aligning RL with Pretraining in Diffusion Models
• Towards a Certificate of Trust: Task-Aware OOD Detection for Scientific AI
• Scaling with Collapse: Efficient and Predictable Training of LLM Families
• ORPO-Distill: Mixed-Policy Preference Optimization for Cross-Architecture LLM Distillation
• Towards generalizable deep ptychography neural networks
• Rethinking Entropy Regularization in Large Reasoning Models
• Learning in an Echo Chamber: Online Learning with Replay Adversary
• BALF: Budgeted Activation-Aware Low-Rank Factorization for Fine-Tuning-Free Model Compression
• High-Dimensional Analysis of Single-Layer Attention for Sparse-Token Classification
• Chance-constrained Flow Matching for High-Fidelity Constraint-aware Generation
• Does Weak-to-strong Generalization Happen under Spurious Correlations?
• SLA: Beyond Sparsity in Diffusion Transformers via Fine-Tunable Sparse-Linear Attention
• Pretraining Scaling Laws for Generative Evaluations of Language Models
• GPS-MTM: Capturing Pattern of Normalcy in GPS-Trajectories with self-supervised learning
• Optimism as Risk-Seeking in Multi-Agent Reinforcement Learning
• Collaborative Device-Cloud LLM Inference through Reinforcement Learning
• On The Variability of Concept Activation Vectors
• In-Context Compositional Q-Learning for Offline Reinforcement Learning
• A Small Math Model: Recasting Strategy Choice Theory in an LLM-Inspired Architecture
• AQUAIR: A High-Resolution Indoor Environmental Quality Dataset for Smart Aquaculture Monitoring
• A Family of Kernelized Matrix Costs for Multiple-Output Mixture Neural Networks
• Demographic-Agnostic Fairness without Harm
• PEARL: Peer-Enhanced Adaptive Radio via On-Device LLM
• Clebsch-Gordan Transformer: Fast and Global Equivariant Attention
• ADAPT: Lightweight, Long-Range Machine Learning Force Fields Without Graphs
• GeoFunFlow: Geometric Function Flow Matching for Inverse Operator Learning over Complex Geometries
• HyMaTE: A Hybrid Mamba and Transformer Model for EHR Representation Learning
• Echo Flow Networks
• The Impossibility of Inverse Permutation Learning in Transformer Models
• A signal separation view of classification
• Evaluation of Machine and Deep Learning Techniques for Cyclone Trajectory Regression and Status Classification by Time Series Data
• Stable Forgetting: Bounded Parameter-Efficient Unlearning in LLMs
• Multi-Scale Geometric Autoencoder
• Model Correlation Detection via Random Selection Probing
• FM-FoG: A Real-Time Foundation Model-based Wearable System for Freezing-of-Gait Mitigation
• Negative Pre-activations Differentiate Syntax
• Group-Relative REINFORCE Is Secretly an Off-Policy Algorithm: Demystifying Some Myths About GRPO and Its Friends
• MDD-Thinker: Towards Large Reasoning Models for Major Depressive Disorder Diagnosis
• Conda: Column-Normalized Adam for Training Large Language Models Faster
• Semantic Editing with Coupled Stochastic Differential Equations
• Proposing a Framework for Machine Learning Adoption on Legacy Systems
• Accessible, Realistic, and Fair Evaluation of Positive-Unlabeled Learning Algorithms
• ChessArena: A Chess Testbed for Evaluating Strategic Reasoning Capabilities of Large Language Models
• Graph Foundation Models: Bridging Language Model Paradigms and Graph Optimization
• Adversarial Reinforcement Learning Framework for ESP Cheater Simulation
• ELASTIQ: EEG-Language Alignment with Semantic Task Instruction and Querying
• Asynchronous Policy Gradient Aggregation for Efficient Distributed Reinforcement Learning
• A study of Universal ODE approaches to predicting soil organic carbon
• Rethinking JEPA: Compute-Efficient Video SSL with Frozen Teachers
• AuON: A Linear-time Alternative to Semi-Orthogonal Momentum Updates
• H+: An Efficient Similarity-Aware Aggregation for Byzantine Resilient Federated Learning
• Towards Generalizable PDE Dynamics Forecasting via Physics-Guided Invariant Learning
• Expanding Horizons of Level Diversity via Multi-objective Evolutionary Learning
• Watermarking Diffusion Language Models
• Evolution Strategies at Scale: LLM Fine-Tuning Beyond Reinforcement Learning
• AXIS: Explainable Time Series Anomaly Detection with Large Language Models
• Muon: Training and Trade-offs with Latent Attention and MoE
• ScatterAD: Temporal-Topological Scattering Mechanism for Time Series Anomaly Detection
• BiHDTrans: binary hyperdimensional transformer for efficient multivariate time series classification
• Semantic Compression via Multimodal Representation Learning
• EOE: Evolutionary Optimization of Experts for Training Language Models
• Distributionally Robust Federated Learning with Outlier Resilience
• Interpretable Kernel Representation Learning at Scale: A Unified Framework Utilizing Nystr\"om Approximation
• FS-KAN: Permutation Equivariant Kolmogorov-Arnold Networks via Function Sharing
• One-Prompt Strikes Back: Sparse Mixture of Experts for Prompt-based Continual Learning
• Guided Uncertainty Learning Using a Post-Hoc Evidential Meta-Model
• LLM DNA: Tracing Model Evolution via Functional Representations
• Specialization after Generalization: Towards Understanding Test-Time Training in Foundation Models
• Trading Carbon for Physics: On the Resource Efficiency of Machine Learning for Spatio-Temporal Forecasting
• LEAF: A Robust Expert-Based Framework for Few-Shot Continual Event Detection
• Training-Free Multimodal Guidance for Video to Audio Generation
• Short window attention enables long-term memorization
• EVO-LRP: Evolutionary Optimization of LRP for Interpretable Model Explanations
• Sketching Low-Rank Plus Diagonal Matrices
• Toward a Holistic Approach to Continual Model Merging
• Avoid Catastrophic Forgetting with Rank-1 Fisher from Diffusion Models
• Characteristic Root Analysis and Regularization for Linear Time Series Forecasting
• GraphIFE: Rethinking Graph Imbalance Node Classification via Invariant Learning
• DRIK: Distribution-Robust Inductive Kriging without Information Leakage
• PreScope: Unleashing the Power of Prefetching for Resource-Constrained MoE Inference
• Virtual Nodes based Heterogeneous Graph Convolutional Neural Network for Efficient Long-Range Information Aggregation
• Pure Node Selection for Imbalanced Graph Node Classification
• Calibration Meets Reality: Making Machine Learning Predictions Trustworthy
• Beyond Greedy Exits: Improved Early Exit Decisions for Risk Control and Reliability
• Why Alignment Must Precede Distillation: A Minimal Working Explanation
• Multi-Scale Spatial-Temporal Hypergraph Network with Lead-Lag Structures for Stock Time Series Forecasting
• Graph Neural Networks with Diversity-aware Neighbor Selection and Dynamic Multi-scale Fusion for Multivariate Time Series Forecasting
• Towards a Comprehensive Scaling Law of Mixture-of-Experts
• Decentralized Dynamic Cooperation of Personalized Models for Federated Continual Learning
• Hedonic Neurons: A Mechanistic Mapping of Latent Coalitions in Transformer MLPs
• FedDAPL: Toward Client-Private Generalization in Federated Learning
• Merge Now, Regret Later: The Hidden Cost of Model Merging is Adversarial Transferability
• Estimating Time Series Foundation Model Transferability via In-Context Learning
• Bridging Discrete and Continuous RL: Stable Deterministic Policy Gradient with Martingale Characterization
• FraudTransformer: Time-Aware GPT for Transaction Fraud Detection
• A Self-Adaptive Frequency Domain Network for Continuous Intraoperative Hypotension Prediction
• GBSK: Skeleton Clustering via Granular-ball Computing and Multi-Sampling for Large-Scale Data
• Time-Shifted Token Scheduling for Symbolic Music Generation
• An Investigation of Batch Normalization in Off-Policy Actor-Critic Algorithms
• Anchored Supervised Fine-Tuning
• SHAPoint: Task-Agnostic, Efficient, and Interpretable Point-Based Risk Scoring via Shapley Values
• Knowledge Homophily in Large Language Models
• Trained Mamba Emulates Online Gradient Descent in In-Context Linear Regression
• Visual CoT Makes VLMs Smarter but More Fragile
• Enhancing LLM Steering through Sparse Autoencoder-Based Vector Refinement
• STAIR: Addressing Stage Misalignment through Temporal-Aligned Preference Reinforcement Learning
• FedAgentBench: Towards Automating Real-world Federated Medical Image Analysis with Server-Client LLM Agents
• Beyond the Exploration-Exploitation Trade-off: A Hidden State Approach for LLM Reasoning in RLVR
• Tequila: Trapping-free Ternary Quantization for Large Language Models
• IndexNet: Timestamp and Variable-Aware Modeling for Time Series Forecasting
• Test-time GNN Model Evaluation on Dynamic Graphs
• Space Group Conditional Flow Matching
• Electric Currents for Discrete Data Generation
• Bayesian Mixture-of-Experts: Towards Making LLMs Know What They Don't Know
• Adversarial Diffusion for Robust Reinforcement Learning
• Efficient Multi-turn RL for GUI Agents via Decoupled Training and Adaptive Data Curation
• Towards Understanding Subliminal Learning: When and How Hidden Biases Transfer
• Gradient Flow Convergence Guarantee for General Neural Network Architectures
• Dynamic Orthogonal Continual Fine-tuning for Mitigating Catastrophic Forgettings
• Differentiable Sparsity via $D$-Gating: Simple and Versatile Structured Penalization
• Integrated Communication and Control for Energy-Efficient UAV Swarms: A Multi-Agent Reinforcement Learning Approach
• Graph Mixing Additive Networks
• HiViS: Hiding Visual Tokens from the Drafter for Speculative Decoding in Vision-Language Models
• Beyond Benchmarks: Understanding Mixture-of-Experts Models through Internal Mechanisms
• Diffusion Models are Kelly Gamblers
• Brain-language fusion enables interactive neural readout and in-silico experimentation
• Efficient Identification of High Similarity Clusters in Polygon Datasets
• Explore-Execute Chain: Towards an Efficient Structured Reasoning Paradigm
• DiBS-MTL: Transformation-Invariant Multitask Learning with Direction Oracles
• Evaluating the Robustness of Chinchilla Compute-Optimal Scaling
• Detecting and Rectifying Noisy Labels: A Similarity-based Approach
• Curriculum-Guided Reinforcement Learning for Synthesizing Gas-Efficient Financial Derivatives Contracts
• Guide: Generalized-Prior and Data Encoders for DAG Estimation
• Drift-Adapter: A Practical Approach to Near Zero-Downtime Embedding Model Upgrades in Vector Databases
• Memory-Efficient Fine-Tuning via Low-Rank Activation Compression
• Statistical Learning Guarantees for Group-Invariant Barron Functions
• Temporal Generalization: A Reality Check
• Revisiting Multivariate Time Series Forecasting with Missing Values
• Beyond Outliers: A Study of Optimizers Under Quantization
• Disentanglement of Variations with Multimodal Generative Modeling
• Fusing Sequence Motifs and Pan-Genomic Features: Antimicrobial Resistance Prediction using an Explainable Lightweight 1D CNN-XGBoost Ensemble
• Improving constraint-based discovery with robust propagation and reliable LLM priors
• RHYTHM: Reasoning with Hierarchical Temporal Tokenization for Human Mobility
• Impute-MACFM: Imputation based on Mask-Aware Flow Matching
• C$^2$GSPG: Confidence-calibrated Group Sequence Policy Gradient towards Self-aware Reasoning
• Trust Region Reward Optimization and Proximal Inverse Reward Optimization Algorithm
• Beyond Heuristics: Globally Optimal Configuration of Implicit Neural Representations
• TimeExpert: Boosting Long Time Series Forecasting with Temporal Mix of Experts
• Critique to Verify: Accurate and Honest Test-Time Scaling with RL-Trained Verifiers
• CrystalGym: A New Benchmark for Materials Discovery Using Reinforcement Learning
• Deep Learning-Based Detection of Cognitive Impairment from Passive Smartphone Sensing with Routine-Aware Augmentation and Demographic Personalization
• ProtoTS: Learning Hierarchical Prototypes for Explainable Time Series Forecasting
• Dense associative memory on the Bures-Wasserstein space
• F-Adapter: Frequency-Adaptive Parameter-Efficient Fine-Tuning in Scientific Machine Learning
• ZeroSiam: An Efficient Siamese for Test-Time Entropy Optimization without Collapse
• CoSIFL: Collaborative Secure and Incentivized Federated Learning with Differential Privacy
• Bridging the Gap Between Promise and Performance for Microscaling FP4 Quantization
• Towards Monotonic Improvement in In-Context Reinforcement Learning
• One-Shot Multi-Label Causal Discovery in High-Dimensional Event Sequences
• WirelessMathLM: Teaching Mathematical Reasoning for LLMs in Wireless Communications with Reinforcement Learning
• SPEC-RL: Accelerating On-Policy Reinforcement Learning via Speculative Rollouts
• More Data or Better Algorithms: Latent Diffusion Augmentation for Deep Imbalanced Regression
• Adaptive Token-Weighted Differential Privacy for LLMs: Not All Tokens Require Equal Protection
• Deep Learning for Subspace Regression
• NanoFlux: Adversarial Dual-LLM Evaluation and Distillation For Multi-Domain Reasoning
• ABConformer: Physics-inspired Sliding Attention for Antibody-Antigen Interface Prediction
• CREPE: Controlling Diffusion with Replica Exchange
• Transfer Learning and Machine Learning for Training Five Year Survival Prognostic Models in Early Breast Cancer
• Continuous-Time Reinforcement Learning for Asset-Liability Management
• A Neural ODE Approach to Aircraft Flight Dynamics Modelling
• ASTGI: Adaptive Spatio-Temporal Graph Interactions for Irregular Multivariate Time Series Forecasting
• Two-Scale Latent Dynamics for Recurrent-Depth Transformers
• MELCOT: A Hybrid Learning Architecture with Marginal Preservation for Matrix-Valued Regression
• LLM Interpretability with Identifiable Temporal-Instantaneous Representation
• Robust Fine-Tuning from Non-Robust Pretrained Models: Mitigating Suboptimal Transfer With Adversarial Scheduling
• Entering the Era of Discrete Diffusion Models: A Benchmark for Schr\"odinger Bridges and Entropic Optimal Transport
• Landing with the Score: Riemannian Optimization through Denoising
• Emergence of Superposition: Unveiling the Training Dynamics of Chain of Continuous Thought
• Splines-Based Feature Importance in Kolmogorov-Arnold Networks: A Framework for Supervised Tabular Data Dimensionality Reduction
• Graph Your Own Prompt
• Planner Aware Path Learning in Diffusion Language Models Training
• Mind the Links: Cross-Layer Attention for Link Prediction in Multiplex Networks
• PATCH: Learnable Tile-level Hybrid Sparsity for LLMs
• URS: A Unified Neural Routing Solver for Cross-Problem Zero-Shot Generalization
• LOTFormer: Doubly-Stochastic Linear Attention via Low-Rank Optimal Transport
• Better Hessians Matter: Studying the Impact of Curvature Approximations in Influence Functions
• Factor Decorrelation Enhanced Data Removal from Deep Predictive Models
• PHASE: Physics-Integrated, Heterogeneity-Aware Surrogates for Scientific Simulations
• Data-Efficient Training by Evolved Sampling
• Generative Evolutionary Meta-Solver (GEMS): Scalable Surrogate-Free Multi-Agent Learning
• Solve Smart, Not Often: Policy Learning for Costly MILP Re-solving
• Localizing Adversarial Attacks To Produces More Imperceptible Noise
• In-Context Learning can Perform Continual Learning Like Humans
• Communication-Efficient and Interoperable Distributed Learning
• On the Capacity of Self-Attention
• Boundary on the Table: Efficient Black-Box Decision-Based Attacks for Structured Data
• Adaptive Margin RLHF via Preference over Preferences
• Observation-Free Attacks on Online Learning to Rank
• Neighborhood Sampling Does Not Learn the Same Graph Neural Network
• From Noise to Knowledge: A Comparative Study of Acoustic Anomaly Detection Models in Pumped-storage Hydropower Plants
• FedCF: Fair Federated Conformal Prediction
• Guided Manifold Alignment with Geometry-Regularized Twin Autoencoders
• Rethinking Large Language Model Distillation: A Constrained Markov Decision Process Perspective
• MonoCon: A general framework for learning ultra-compact high-fidelity representations using monotonicity constraints
• Compute-Optimal Quantization-Aware Training
• Understanding SOAP from the Perspective of Gradient Whitening
• SINQ: Sinkhorn-Normalized Quantization for Calibration-Free Low-Precision LLM Weights
• Meta-Learning Fourier Neural Operators for Hessian Inversion and Enhanced Variational Data Assimilation
• GDR-learners: Orthogonal Learning of Generative Models for Potential Outcomes
• Doubly-Robust LLM-as-a-Judge: Externally Valid Estimation with Imperfect Personas
• Reinforcement Learning with Discrete Diffusion Policies for Combinatorial Action Spaces
• Functional Critic Modeling for Provably Convergent Off-Policy Actor-Critic
• Shape-Informed Clustering of Multi-Dimensional Functional Data via Deep Functional Autoencoders
• OptiMind: Teaching LLMs to Think Like Optimization Experts
• MDP modeling for multi-stage stochastic programs
• T-TAMER: Provably Taming Trade-offs in ML Serving
• Analysis of Variational Autoencoders
• Sample-efficient Multiclass Calibration under $\ell_{p}$ Error
• Physically Plausible Multi-System Trajectory Generation and Symmetry Discovery
• MoE-PHDS: One MoE checkpoint for flexible runtime sparsity
• On the Sheafification of Higher-Order Message Passing
• Tracing the Representation Geometry of Language Models from Pretraining to Post-training
• Understanding Catastrophic Interference On the Identifibility of Latent Representations
• DPFNAS: Differential Privacy-Enhanced Federated Neural Architecture Search for 6G Edge Intelligence
• GuardNet: Graph-Attention Filtering for Jailbreak Defense in Large Language Models
• IsingFormer: Augmenting Parallel Tempering With Learned Proposals
• Beyond Aggregation: Guiding Clients in Heterogeneous Federated Learning
• Understanding Language Prior of LVLMs by Contrasting Chain-of-Embedding
• Dynamics of Learning: Generative Schedules from Latent ODEs
• Beyond Model Ranking: Predictability-Aligned Evaluation for Time Series Forecasting
• CLAD-Net: Continual Activity Recognition in Multi-Sensor Wearable Systems
• Signal Preserving Weight Initialization for Odd-Sigmoid Activations
• Unleashing Flow Policies with Distributional Critics
• Demystifying Network Foundation Models
• Sensitivity Analysis for Diffusion Models
• Causally-Enhanced Reinforcement Policy Optimization
• Towards Quantum-Ready Blockchain Fraud Detection via Ensemble Graph Neural Networks
• Effective Quantization of Muon Optimizer States