AI Research News Feeds for August 11th, 2025

AI RESEARCH PAPERS & ACADEMIC SOURCES

• Identifiability of the minimum-trace directed acyclic graph and hill climbing algorithms without strict local optima under weakly increasing error variances
• Estimating the size of a set using cascading exclusion
• A variational approach to dimension-free self-normalized concentration
• Non-asymptotic estimates for accelerated high order Langevin Monte Carlo algorithms
• Proximal optimal transport divergences
• Can Multimodal Large Language Models Understand Spatial Relations?
• TD3Net: A Temporal Densely Connected Multi-Dilated Convolutional Network for Lipreading
• End-to-End Fine-Tuning of 3D Texture Generation using Differentiable Rewards
• Neural-Driven Image Editing
• Trustworthy Pedestrian Trajectory Prediction via Pattern-Aware Interaction Modeling
• InterAct-Video: Reasoning-Rich Video QA for Urban Traffic
• Survival Modeling from Whole Slide Images via Patch-Level Graph Clustering and Mixture Density Experts
• Direct Robot Configuration Space Construction using Convolutional Encoder-Decoders
• MambaEviScrib: Mamba and Evidence-Guided Consistency Enhance CNN Robustness for Scribble-Based Weakly Supervised Ultrasound Image Segmentation
• CDI: Blind Image Restoration Fidelity Evaluation based on Consistency with Degraded Image
• POMATO: Marrying Pointmap Matching with Temporal Motion for Dynamic 3D Reconstruction
• DAVSP: Safety Alignment for Large Vision-Language Models via Deep Aligned Visual Safety Prompt
• MoDA: Multi-modal Diffusion Architecture for Talking Head Generation
• PASG: A Closed-Loop Framework for Automated Geometric Primitive Extraction and Semantic Anchoring in Robotic Manipulation
• AnimateScene: Camera-controllable Animation in Any Scene
• Fast Motion Estimation and Context-Aware Refinement for Efficient Bayer-Domain Video Vision
• EvoMakeup: High-Fidelity and Controllable Makeup Editing with MakeupQuad
• MathReal: We Keep It Real! A Real Scene Benchmark for Evaluating Math Reasoning in Multimodal Large Language Models
• ExploreGS: Explorable 3D Scene Reconstruction with Virtual Camera Samplings and Diffusion Priors
• Learning 3D Texture-Aware Representations for Parsing Diverse Human Clothing and Body Parts
• InstantEdit: Text-Guided Few-Step Image Editing with Piecewise Rectified Flow
• More Is Better: A MoE-Based Emotion Recognition Framework with Human Preference Alignment
• NEP: Autoregressive Image Editing via Next Editing Token Prediction
• VQAThinker: Exploring Generalizable and Explainable Video Quality Assessment via Reinforcement Learning
• LV-Net: Anatomy-aware lateral ventricle shape modeling with a case study on Alzheimer's disease, the Australian Imaging Biomarkers and Lifestyle flagship study of ageing
• Lightweight Quad Bayer HybridEVS Demosaicing via State Space Augmented Cross-Attention
• Distribution-Specific Learning for Joint Salient and Camouflaged Object Detection
• DreamVE: Unified Instruction-based Image and Video Editing
• SwiftVideo: A Unified Framework for Few-Step Video Generation through Trajectory-Distribution Alignment
• AdaptInfer: Adaptive Token Pruning for Vision-Language Model Inference with Dynamical Text Guidance
• Q-CLIP: Unleashing the Power of Vision-Language Models for Video Quality Assessment through Unified Cross-Modal Adaptation
• E-React: Towards Emotionally Controlled Synthesis of Human Reactions
• UGD-IML: A Unified Generative Diffusion-based Framework for Constrained and Unconstrained Image Manipulation Localization
• MCA: 2D-3D Retrieval with Noisy Labels via Multi-level Adaptive Correction and Alignment
• GMF-Drive: Gated Mamba Fusion with Spatial-Aware BEV Representation for End-to-End Autonomous Driving
• SynSeg: Feature Synergy for Multi-Category Contrastive Learning in Open-Vocabulary Semantic Segmentation
• Learning Representations of Satellite Images with Evaluations on Synoptic Weather Events
• SC-Captioner: Improving Image Captioning with Self-Correction by Reinforcement Learning
• SAM Encoder Breach by Adversarial Simplicial Complex Triggers Downstream Model Failures
• DiffCap: Diffusion-based Real-time Human Motion Capture using Sparse IMUs and a Monocular Camera
• SDEval: Safety Dynamic Evaluation for Multimodal Large Language Models
• Text-guided Visual Prompt DINO for Generic Segmentation
• DSConv: Dynamic Splitting Convolution for Pansharpening
• VISTAR:A User-Centric and Role-Driven Benchmark for Text-to-Image Evaluation
• An Interpretable Multi-Plane Fusion Framework With Kolmogorov-Arnold Network Guided Attention Enhancement for Alzheimer's Disease Diagnosis
• Fewer Denoising Steps or Cheaper Per-Step Inference: Towards Compute-Optimal Diffusion Model Deployment
• Graph-based Robot Localization Using a Graph Neural Network with a Floor Camera and a Feature Rich Industrial Floor
• MA-CBP: A Criminal Behavior Prediction Framework Based on Multi-Agent Asynchronous Collaboration
• A Semantic Segmentation Algorithm for Pleural Effusion Based on DBIF-AUNet
• AnomalyMoE: Towards a Language-free Generalist Model for Unified Visual Anomaly Detection
• PA-HOI: A Physics-Aware Human and Object Interaction Dataset
• Interpretable Rheumatoid Arthritis Scoring via Anatomy-aware Multiple Instance Learning
• TEFormer: Texture-Aware and Edge-Guided Transformer for Semantic Segmentation of Urban Remote Sensing Images
• Depth Jitter: Seeing through the Depth
• Towards Unified Image Deblurring using a Mixture-of-Experts Decoder
• Deepfake Detection that Generalizes Across Benchmarks
• FedX: Explanation-Guided Pruning for Communication-Efficient Federated Learning in Remote Sensing
• XAG-Net: A Cross-Slice Attention and Skip Gating Network for 2.5D Femur MRI Segmentation
• Uncertainty-quantified Rollout Policy Adaptation for Unlabelled Cross-domain Temporal Grounding
• Can Diffusion Models Bridge the Domain Gap in Cardiac MR Imaging?
• ViPro-2: Unsupervised State Estimation via Integrated Dynamics for Guiding Video Prediction
• Street View Sociability: Interpretable Analysis of Urban Social Behavior Across 15 Cities
• Aligning Effective Tokens with Video Anomaly in Large Language Models
• An Implemention of Two-Phase Image Segmentation using the Split Bregman Method
• Text as Any-Modality for Zero-Shot Classification by Consistent Prompt Tuning
• FVGen: Accelerating Novel-View Synthesis with Adversarial Video Diffusion Distillation
• Feature-Space Oversampling for Addressing Class Imbalance in SAR Ship Classification
• MotionSwap
• LightSwitch: Multi-view Relighting with Material-guided Diffusion
• Neural Field-Based 3D Surface Reconstruction of Microstructures from Multi-Detector Signals in Scanning Electron Microscopy
• Universally Unfiltered and Unseen:Input-Agnostic Multimodal Jailbreaks against Text-to-Image Model Safeguards
• KnapFormer: An Online Load Balancer for Efficient Diffusion Transformers Training
• Transformer-Based Explainable Deep Learning for Breast Cancer Detection in Mammography: The MammoFormer Framework
• Clinically-guided Data Synthesis for Laryngeal Lesion Detection
• Affordance-R1: Reinforcement Learning for Generalizable Affordance Reasoning in Multimodal Large Language Model
• Anti-Tamper Protection for Unauthorized Individual Image Generation
• CPT-Interp: Continuous sPatial and Temporal Motion Modeling for 4D Medical Image Interpolation
• A Calibration Tool for Refractive Underwater Vision
• Hybrid-TTA: Continual Test-time Adaptation via Dynamic Domain Shift Detection
• Localized Gaussians as Self-Attention Weights for Point Clouds Correspondence
• MBA-SLAM: Motion Blur Aware Gaussian Splatting SLAM
• SAR Strikes Back: A New Hope for RSVQA
• Fine-Grained Image-Text Correspondence with Cost Aggregation for Open-Vocabulary Part Segmentation
• MPG-SAM 2: Adapting SAM 2 with Mask Priors and Global Context for Referring Video Object Segmentation
• Embodied Intelligence for 3D Understanding: A Survey on 3D Scene Question Answering
• Generative Video Bi-flow
• COIN: Confidence Score-Guided Distillation for Annotation-Free Cell Segmentation
• Can Test-Time Scaling Improve World Foundation Model?
• Two-stage deep learning framework for the restoration of incomplete-ring PET images
• Event2Vec: Processing neuromorphic events directly by representations in vector space
• Dome-DETR: DETR with Density-Oriented Feature-Query Manipulation for Efficient Tiny Object Detection
• Spectrum Projection Score: Aligning Retrieved Summaries with Reader Models in Retrieval-Augmented Generation
• Adversarial Topic-aware Prompt-tuning for Cross-topic Automated Essay Scoring
• ConlangCrafter: Constructing Languages with a Multi-Hop LLM Pipeline
• Few-Shot Prompting for Extractive Quranic QA with Instruction-Tuned LLMs
• You Don't Need Pre-built Graphs for RAG: Retrieval Augmented Generation with Adaptive Reasoning Structures
• AURA: Affordance-Understanding and Risk-aware Alignment Technique for Large Language Models
• Scaling Personality Control in LLMs with Big Five Scaler Prompts
• Semantic and Structural Analysis of Implicit Biases in Large Language Models: An Interpretable Approach
• Pragmatics beyond humans: meaning, communication, and LLMs
• Comparing Knowledge Injection Methods for LLMs in a Low-Resource Regime
• DKG-LLM : A Framework for Medical Diagnosis and Personalized Treatment Recommendations via Dynamic Knowledge Graph and Large Language Model Integration
• Beyond Uniform Criteria: Scenario-Adaptive Multi-Dimensional Jailbreak Evaluation
• EICAP: Deep Dive in Assessment and Enhancement of Large Language Models in Emotional Intelligence through Multi-Turn Conversations
• Matrix-Driven Instant Review: Confident Detection and Reconstruction of LLM Plagiarism on PC
• Cyberbullying Detection via Aggression-Enhanced Prompting
• Evaluating Style-Personalized Text Generation: Challenges and Directions
• LLMs vs. Chinese Anime Enthusiasts: A Comparative Study on Emotionally Supportive Role-Playing
• Quantifying Conversation Drift in MCP via Latent Polytope
• SlimInfer: Accelerating Long-Context LLM Inference via Dynamic Token Pruning
• GLM-4.5: Agentic, Reasoning, and Coding (ARC) Foundation Models
• HapticLLaMA: A Multimodal Sensory Language Model for Haptic Captioning
• DINA: A Dual Defense Framework Against Internal Noise and External Attacks in Natural Language Processing
• Basic interactive algorithms: Preview
• NanoCodec: Towards High-Quality Ultra Fast Speech LLM Inference
• Fact2Fiction: Targeted Poisoning Attack to Agentic Fact-checking System
• Effective Training Data Synthesis for Improving MLLM Chart Understanding
• Benchmarking LLMs on the Semantic Overlap Summarization Task
• Towards Pareto Optimal Throughput in Small Language Model Serving
• Extract-and-Abstract: Unifying Extractive and Abstractive Summarization within Single Encoder-Decoder Framework
• Turning Logic Against Itself : Probing Model Defenses Through Contrastive Questions
• Structural Embedding Projection for Contextual Large Language Model Inference
• Context-Preserving Tensorial Reconfiguration in Large Language Model Training
• Contextual Morphogenesis in Large Language Models: A Novel Approach to Self-Organizing Token Representations
• Context-Aware Hierarchical Merging for Long Document Summarization
• Gradient-Regularized Latent Space Modulation in Large Language Models for Structured Contextual Synthesis
• Latent Structure Modulation in Large Language Models Through Stochastic Concept Embedding Transitions
• Structural Perturbation in Large Language Model Representations through Recursive Symbolic Regeneration
• Structural Reformation of Large Language Model Neuron Encapsulation for Divergent Information Aggregation
• Structured Convergence in Large Language Model Representations via Hierarchical Latent Space Folding
• Statistical Coherence Alignment for Large Language Model Representation Learning Through Tensor Field Convergence
• Exploring Contextual Flux in Large Language Models: A Novel Approach to Self-Modulating Semantic Networks
• One ruler to measure them all: Benchmarking multilingual long-context language models
• OpenCodeReasoning: Advancing Data Distillation for Competitive Coding
• Single-Pass Document Scanning for Question Answering
• Not All Data Are Unlearned Equally
• EvidenceBench: A Benchmark for Extracting Evidence from Biomedical Papers
• Can a Crow Hatch a Falcon? Lineage Matters in Predicting Large Language Model Performance
• The Devil Is in the Word Alignment Details: On Translation-Based Cross-Lingual Transfer for Token Classification Tasks
• Automated Privacy Information Annotation in Large Language Model Interactions
• DrVoice: Parallel Speech-Text Voice Conversation Model via Dual-Resolution Speech Representations
• AALC: Large Language Model Efficient Reasoning via Adaptive Accuracy-Length Control
• Integrating large language models and active inference to understand eye movements in reading and dyslexia
• Reducibility among NP-Hard graph problems and boundary classes
• OpenCodeInstruct: A Large-scale Instruction Tuning Dataset for Code LLMs
• From Autonomy to Agency: Agentic Vehicles for Human-Centered Mobility Systems
• Nyay-Darpan: Enhancing Decision Making Through Summarization and Case Retrieval for Consumer Law in India
• Generalized Few-Shot Out-of-Distribution Detection
• Improving Masked Style Transfer using Blended Partial Convolution
• MAISI-v2: Accelerated 3D High-Resolution Medical Image Synthesis with Rectified Flow and Region-specific Contrastive Loss
• Optimization-Free Style Transfer for 3D Gaussian Splats
• MZEN: Multi-Zoom Enhanced NeRF for 3-D Reconstruction with Unknown Camera Poses
• TSMS-SAM2: Multi-scale Temporal Sampling Augmentation and Memory-Splitting Pruning for Promptable Video Object Segmentation and Tracking in Surgical Scenarios
• Temporal Cluster Assignment for Efficient Real-Time Video Segmentation
• VISTA: Vision-Language Imitation of Situational Thinking and Attention for Human-Like Driver Focus in Dynamic Environments
• Multi-view Gaze Target Estimation
• ETTA: Efficient Test-Time Adaptation for Vision-Language Models through Dynamic Embedding Updates
• HOLODECK 2.0: Vision-Language-Guided 3D World Generation with Editing
• Robust Image Stitching with Optimal Plane
• Neural Field Representations of Mobile Computational Photography
• PEACH: A sentence-aligned Parallel English-Arabic Corpus for Healthcare
• Guardians and Offenders: A Survey on Harmful Content Generation and Safety Mitigation
• FineDialFact: A benchmark for Fine-grained Dialogue Fact Verification
• Human-like fleeting memory improves language learning but impairs reading time prediction in transformer language models
• "Mirror" Language AI Models of Depression are Criterion-Contaminated
• Discovering Properties of Inflectional Morphology in Neural Emergent Communication
• CANVAS: Commonsense-Aware Navigation System for Intuitive Human-Robot Interaction
• WildSAT: Learning Satellite Image Representations from Wildlife Observations
• EVA-S2PLoR: Decentralized Secure 2-party Logistic Regression with A Subtly Hadamard Product Protocol (Full Version)
• Conditional Diffusion Models are Medical Image Classifiers that Provide Explainability and Uncertainty for Free
• Building Age Estimation: A New Multi-Modal Benchmark Dataset and Community Challenge
• Rank1: Test-Time Compute for Reranking in Information Retrieval
• CRUST-Bench: A Comprehensive Benchmark for C-to-safe-Rust Transpilation
• MAATS: A Multi-Agent Automated Translation System Based on MQM Evaluation
• Decompositional Reasoning for Graph Retrieval with Large Language Models
• Measurement as Bricolage: Examining How Data Scientists Construct Target Variables for Predictive Modeling Tasks
• The Fourth State: Signed-Zero Ternary for Stable LLM Quantization (and More)
• Dual Signal Decomposition of Stochastic Time Series
• Fast, Convex and Conditioned Network for Multi-Fidelity Vectors and Stiff Univariate Differential Equations
• Mitigating Think-Answer Mismatch in LLM Reasoning Through Noise-Aware Advantage Reweighting
• LinguaFluid: Language Guided Fluid Control via Semantic Rewards in Reinforcement Learning
• Parameter-free Optimal Rates for Nonlinear Semi-Norm Contractions with Applications to $Q$-Learning
• Pruning the Unsurprising: Efficient Code Reasoning via First-Token Surprisal
• Optimizing Prompt Sequences using Monte Carlo Tree Search for LLM-Based Optimization
• Stepwise Fine and Gray: Subject-Specific Variable Selection Shows When Hemodynamic Data Improves Prognostication of Comatose Post-Cardiac Arrest Patients
• Recurrent Deep Differentiable Logic Gate Networks
• Improving Diagnostic Accuracy for Oral Cancer with inpainting Synthesis Lesions Generated Using Diffusion Models
• SCAR: State-Space Compression for AI-Driven Resource Management in 6G-Enabled Vehicular Infotainment Systems
• Near-Optimal Regret for Efficient Stochastic Combinatorial Semi-Bandits
• Multi-Omics Analysis for Cancer Subtype Inference via Unrolling Graph Smoothness Priors
• A Study on Regularization-Based Continual Learning Methods for Indic ASR
• Low-Bit Data Processing Using Multiple-Output Spiking Neurons with Non-linear Reset Feedback
• Introducing Fractional Classification Loss for Robust Learning with Noisy Labels
• Geometric-k-means: A Bound Free Approach to Fast and Eco-Friendly k-means
• A New Lens on Homelessness: Daily Tent Monitoring with 311 Calls and Street Images
• Sample-efficient LLM Optimization with Reset Replay
• LLM Unlearning using Gradient Ratio-Based Influence Estimation and Noise Injection
• Indian Legal NLP Benchmarks : A Survey
• Moment Estimate and Variational Approach for Learning Generalized Diffusion with Non-gradient Structures
• AI Guided Accelerator For Search Experience
• Random Walk Learning and the Pac-Man Attack
• Domain-Specific Fine-Tuning and Prompt-Based Learning: A Comparative Study for developing Natural Language-Based BIM Information Retrieval Systems
• MM-FusionNet: Context-Aware Dynamic Fusion for Multi-modal Fake News Detection with Large Vision-Language Models
• Boosting Adversarial Transferability via Residual Perturbation Attack
• Leveraging large language models for SQL behavior-based database intrusion detection
• MambaITD: An Efficient Cross-Modal Mamba Network for Insider Threat Detection
• G-UBS: Towards Robust Understanding of Implicit Feedback via Group-Aware User Behavior Simulation
• Reduction Techniques for Survival Analysis
• Detecting Model Misspecification in Cosmology with Scale-Dependent Normalizing Flows
• Evaluating Universal Machine Learning Force Fields Against Experimental Measurements
• Stochastic Trace Optimization of Parameter Dependent Matrices Based on Statistical Learning Theory
• Stochastic Bandits for Crowdsourcing and Multi-Platform Autobidding
• Training chord recognition models on artificially generated audio
• Hybrid Physics-Machine Learning Models for Quantitative Electron Diffraction Refinements
• Enhancing Construction Site Analysis and Understanding with 3D Segmentation
• Position: Intelligent Coding Systems Should Write Programs with Justifications
• Efficient Knowledge Probing of Large Language Models by Adapting Pre-trained Embeddings
• Data-Driven Density Steering via the Gromov-Wasserstein Optimal Transport Distance
• AGI for the Earth, the path, possibilities and how to evaluate intelligence of models that work with Earth Observation Data?
• Lightweight Auto-bidding based on Traffic Prediction in Live Advertising
• Adaptive Backtracking for Privacy Protection in Large Language Models
• Ensemble-Based Graph Representation of fMRI Data for Cognitive Brain State Classification
• IOCC: Aligning Semantic and Cluster Centers for Few-shot Short Text Clustering
• Enhancing the Scalability of Classical Surrogates for Real-World Quantum Machine Learning Applications
• Large Language Model Data Generation for Enhanced Intent Recognition in German Speech
• EmoAugNet: A Signal-Augmented Hybrid CNN-LSTM Framework for Speech Emotion Recognition
• Decorrelated feature importance from local sample weighting
• DP-SPRT: Differentially Private Sequential Probability Ratio Tests
• Tree-Based Deep Learning for Ranking Symbolic Integration Algorithms
• Blockchain-Enabled Federated Learning
• eSASRec: Enhancing Transformer-based Recommendations in a Modular Fashion
• TRUST: Leveraging Text Robustness for Unsupervised Domain Adaptation
• Maximum Impact with Fewer Features: Efficient Feature Selection for Cold-Start Recommenders through Collaborative Importance Weighting
• Multivariate Fields of Experts
• Reinforcement Learning Based Sensor Optimization for Bio-markers
• Bayesian Gaussian Process ODEs via Double Normalizing Flows
• Decision-focused predictions via pessimistic bilevel optimization: complexity and algorithms
• Soft Dice Confidence: A Near-Optimal Confidence Estimator for Selective Prediction in Semantic Segmentation
• Data Collaboration Analysis with Orthonormal Basis Selection and Alignment
• Position: Lifetime tuning is incompatible with continual reinforcement learning
• Formal Local Implication Between Two Neural Networks
• Adaptive Collocation Point Strategies For Physics Informed Neural Networks via the QR Discrete Empirical Interpolation Method
• Rethinking the Bias of Foundation Model under Long-tailed Distribution
• The Ensemble Kalman Update is an Empirical Matheron Update
• Sample-Efficient Reinforcement Learning from Human Feedback via Information-Directed Sampling
• Navigating Demand Uncertainty in Container Shipping: Deep Reinforcement Learning for Enabling Adaptive and Feasible Master Stowage Planning
• Global graph features unveiled by unsupervised geometric deep learning
• Learning to Match Unpaired Data with Minimum Entropy Coupling
• Echo Chamber: RL Post-training Amplifies Behaviors Learned in Pretraining
• On the Value of Cross-Modal Misalignment in Multimodal Representation Learning
• Khan-GCL: Kolmogorov-Arnold Network Based Graph Contrastive Learning with Hard Negatives
• A Graph Sufficiency Perspective for Neural Networks
• MESAHA-Net: Multi-Encoders based Self-Adaptive Hard Attention Network with Maximum Intensity Projections for Lung Nodule Segmentation in CT Scan
• Generalization Bound for Diffusion Models using Random Features
• Towards More Realistic Extraction Attacks: An Adversarial Perspective
• Modeling Spatial Extremal Dependence of Precipitation Using Distributional Neural Networks
• Optimal sampling for least-squares approximation
• Optimal Linear Baseline Models for Scientific Machine Learning
• An Effective Approach for Node Classification in Textual Graphs
• A Markov Decision Process Framework for Early Maneuver Decisions in Satellite Collision Avoidance
• Graph Federated Learning for Personalized Privacy Recommendation
• Reparameterization Proximal Policy Optimization
• InfoCausalQA:Can Models Perform Non-explicit Causal Reasoning Based on Infographic?
• Membership Inference Attack with Partial Features
• In-Training Defenses against Emergent Misalignment in Language Models
• Synthetic Data Generation and Differential Privacy using Tensor Networks' Matrix Product States (MPS)
• SIFThinker: Spatially-Aware Image Focus for Visual Reasoning
• Numerical Considerations in Weighted Model Counting
• OM2P: Offline Multi-Agent Mean-Flow Policy
• Advanced Deep Learning Techniques for Accurate Lung Cancer Detection and Classification
• FedMeNF: Privacy-Preserving Federated Meta-Learning for Neural Fields
• Mixture of Experts Guided by Gaussian Splatters Matters: A new Approach to Weakly-Supervised Video Anomaly Detection
• Unsupervised Partner Design Enables Robust Ad-hoc Teamwork
• On Approximate MMS Allocations on Restricted Graph Classes
• Harnessing Adaptive Topology Representations for Zero-Shot Graph Question Answering
• Structural Equation-VAE: Disentangled Latent Representations for Tabular Data
• Are you In or Out (of gallery)? Wisdom from the Same-Identity Crowd
• Beyond Prompt-Induced Lies: Investigating LLM Deception on Benign Prompts
• ActivityDiff: A diffusion model with Positive and Negative Activity Guidance for De Novo Drug Design
• SpeakerLM: End-to-End Versatile Speaker Diarization and Recognition with Multimodal Large Language Models
• End-to-End Text-to-SQL with Dataset Selection: Leveraging LLMs for Adaptive Query Generation
• Identity Increases Stability in Neural Cellular Automata
• Robust Target Speaker Diarization and Separation via Augmented Speaker Embedding Sampling
• A Systematic Literature Review of Retrieval-Augmented Generation: Techniques, Metrics, and Challenges
• A Classification-Aware Super-Resolution Framework for Ship Targets in SAR Imagery
• Dimensional Characterization and Pathway Modeling for Catastrophic AI Risks
• Shortcut Learning in Generalist Robot Policies: The Role of Dataset Diversity and Fragmentation
• SPARSE Data, Rich Results: Few-Shot Semi-Supervised Learning via Class-Conditioned Image Translation
• Memp: Exploring Agent Procedural Memory
• CLIPin: A Non-contrastive Plug-in to CLIP for Multimodal Semantic Alignment
• Learning the Topic, Not the Language: How LLMs Classify Online Immigration Discourse Across Languages
• Echoes of Automation: The Increasing Use of LLMs in Newsmaking
• Text Embedded Swin-UMamba for DeepLesion Segmentation
• ScamAgents: How AI Agents Can Simulate Human-Level Scam Calls
• Intuition emerges in Maximum Caliber models at criticality
• Post-training for Efficient Communication via Convention Formation
• WGAST: Weakly-Supervised Generative Network for Daily 10 m Land Surface Temperature Estimation via Spatio-Temporal Fusion
• From Next-Token to Mathematics: The Learning Dynamics of Mathematical Reasoning in Language Models
• Are Your LLMs Capable of Stable Reasoning?
• Probabilistic Foundations for Metacognition via Hybrid-AI
• Hypothesis-Driven Theory-of-Mind Reasoning for Large Language Models
• Off-Policy Evaluation for Sequential Persuasion Process with Unobserved Confounding
• DONOD: Efficient and Generalizable Instruction Fine-Tuning for LLMs via Model-Intrinsic Dataset Pruning
• Contemplative Artificial Intelligence
• Language Agents Mirror Human Causal Reasoning Biases. How Can We Help Them Think Like Scientists?
• SuperRL: Reinforcement Learning with Supervision to Boost Language Model Reasoning
• HASD: Hierarchical Adaption for pathology Slide-level Domain-shift
• Efficient Knowledge Graph Construction and Retrieval from Unstructured Text for Large-Scale RAG Systems
• Benchmarking Deception Probes via Black-to-White Performance Boosts
• From research to clinic: Accelerating the translation of clinical decision support systems by making synthetic data interoperable
• A Markov Random Field model for Hypergraph-based Machine Learning
• Learning to Initialize Trajectory Optimization for Vision-Based Autonomous Flight in Unknown Environments
• Improved DDIM Sampling with Moment Matching Gaussian Mixtures
• Entropy Causal Graphs for Multivariate Time Series Anomaly Detection
• INS-MMBench: A Comprehensive Benchmark for Evaluating LVLMs' Performance in Insurance
• Reorganizing attention-space geometry with expressive attention
• Spatio-Temporal Partial Sensing Forecast for Long-term Traffic
• ATM: Improving Model Merging by Alternating Tuning and Merging
• Reconsidering the Performance of GAE in Link Prediction
• CodeXEmbed: A Generalist Embedding Model Family for Multiligual and Multi-task Code Retrieval
• DeepMDV: Global Spatial Matching for Multi-depot Vehicle Routing Problems
• TryOffDiff: Virtual-Try-Off via High-Fidelity Garment Reconstruction using Diffusion Models
• LeakAgent: RL-based Red-teaming Agent for LLM Privacy Leakage
• The Alternative Annotator Test for LLM-as-a-Judge: How to Statistically Justify Replacing Human Annotators with LLMs
• Neural Contextual Reinforcement Framework for Logical Structure Language Generation
• Unveiling Zero-Space Detection: A Novel Framework for Autonomous Ransomware Identification in High-Velocity Environments
• Architectural Fusion Through Contextual Partitioning in Large Language Models: A Novel Approach to Parameterized Knowledge Integration
• Autonomous Structural Memory Manipulation for Large Language Models Using Hierarchical Embedding Augmentation
• Systemizing Multiplicity: The Curious Case of Arbitrariness in Machine Learning
• Hierarchical Pattern Decryption Methodology for Ransomware Detection Using Probabilistic Cryptographic Footprints
• FIT-Print: Towards False-claim-resistant Model Ownership Verification via Targeted Fingerprint
• Contextual Reinforcement in Multimodal Token Compression for Large Language Models
• Algorithmic Segmentation and Behavioral Profiling for Ransomware Detection Using Temporal-Correlation Graphs
• Contextually Entangled Gradient Mapping for Optimized LLM Comprehension
• DeToNATION: Decoupled Torch Network-Aware Training on Interlinked Online Nodes
• CAST: Cross Attention based multimodal fusion of Structure and Text for materials property prediction
• ArchRAG: Attributed Community-based Hierarchical Retrieval-Augmented Generation
• Exploring Synaptic Resonance in Large Language Models: A Novel Approach to Contextual Memory Integration
• Topic Over Source: The Key to Effective Data Mixing for Language Models Pre-training
• ACTIVA: Amortized Causal Effect Estimation via Transformer-based Variational Autoencoder
• Code-as-Symbolic-Planner: Foundation Model-Based Robot Planning via Symbolic Code Generation
• Training Plug-n-Play Knowledge Modules with Deep Context Distillation
• CodeARC: Benchmarking Reasoning Capabilities of LLM Agents for Inductive Program Synthesis
• Evaluating and Designing Sparse Autoencoders by Approximating Quasi-Orthogonality
• M$^2$IV: Towards Efficient and Fine-grained Multimodal In-Context Learning via Representation Engineering
• Self-Steering Language Models
• Layers at Similar Depths Generate Similar Activations Across LLM Architectures
• AI-Assisted Conversational Interviewing: Effects on Data Quality and User Experience
• iTFKAN: Interpretable Time Series Forecasting with Kolmogorov-Arnold Network
• SMOGAN: Synthetic Minority Oversampling with GAN Refinement for Imbalanced Regression
• Vision-Language Model-Based Semantic-Guided Imaging Biomarker for Lung Nodule Malignancy Prediction
• Solving Copyright Infringement on Short Video Platforms: Novel Datasets and an Audio Restoration Deep Learning Pipeline
• No Query, No Access
• Are Large Language Models Robust in Understanding Code Against Semantics-Preserving Mutations?
• LaDi-WM: A Latent Diffusion-based World Model for Predictive Manipulation
• CUB: Benchmarking Context Utilisation Techniques for Language Models
• LLM-Meta-SR: In-Context Learning for Evolving Selection Operators in Symbolic Regression
• Fusing Cross-Domain Knowledge from Multimodal Data to Solve Problems in the Physical World
• Survey on the Evaluation of Generative Models in Music
• AVA-Bench: Atomic Visual Ability Benchmark for Vision Foundation Models
• No Universal Prompt: Unifying Reasoning through Adaptive Prompting for Temporal Table Reasoning
• Scientifically-Interpretable Reasoning Network (ScIReN): Discovering Hidden Relationships in the Carbon Cycle and Beyond
• Time-Prompt: Integrated Heterogeneous Prompts for Unlocking LLMs in Time Series Forecasting
• Quantifying Fairness in LLMs Beyond Tokens: A Semantic and Statistical Perspective
• Quality over Quantity: An Effective Large-Scale Data Reduction Strategy Based on Pointwise V-Information
• Humans overrely on overconfident language models, across languages
• Diagrams-to-Dynamics (D2D): Exploring Causal Loop Diagram Leverage Points under Uncertainty
• A Graph Neural Network Approach for Mapping the Conceptual Structure and Inter-Branch Connectivity of Physics
• Machine Learning-Based Nonlinear Nudging for Chaotic Dynamical Systems
• InfiGUI-G1: Advancing GUI Grounding with Adaptive Exploration Policy Optimization
• A Framework for Inherently Safer AGI through Language-Mediated Active Inference
• Whither symbols in the era of advanced neural networks?
• Holistic Explainable AI (H-XAI): Extending Transparency Beyond Developers in AI-Driven Decision Making
• Safety of Embodied Navigation: A Survey
• Planning Agents on an Ego-Trip: Leveraging Hybrid Ego-Graph Ensembles for Improved Tool Retrieval in Enterprise Task Planning
• Mediator-Guided Multi-Agent Collaboration among Open-Source Models for Medical Decision-Making
• Society of Mind Meets Real-Time Strategy: A Hierarchical Multi-Agent Framework for Strategic Reasoning
• LLMs for Resource Allocation: A Participatory Budgeting Approach to Inferring Preferences
• Don't Forget Imagination!
• A Generic Complete Anytime Beam Search for Optimal Decision Tree
• ME$^3$-BEV: Mamba-Enhanced Deep Reinforcement Learning for End-to-End Autonomous Driving with BEV-Perception
• Aggregate-Combine-Readout GNNs Are More Expressive Than Logic C2
• PanelTR: Zero-Shot Table Reasoning Framework Through Multi-Agent Scientific Discussion
• SKATE, a Scalable Tournament Eval: Weaker LLMs differentiate between stronger ones using verifiable challenges
• Study of Robust Features in Formulating Guidance for Heuristic Algorithms for Solving the Vehicle Routing Problem
• Retrieval Augmented Large Language Model System for Comprehensive Drug Contraindications
• Overconfidence in LLM-as-a-Judge: Diagnosis and Confidence-Driven Solution
• GeoLaux: A Benchmark for Evaluating MLLMs' Geometry Performance on Long-Step Problems Requiring Auxiliary Lines
• Learning Logical Rules using Minimum Message Length
• Symmetry breaking for inductive logic programming
• LLM Robustness Leaderboard v1 --Technical report
• A "good regulator theorem" for embodied agents
• AntiCheatPT: A Transformer-Based Approach to Cheat Detection in Competitive Computer Games
• From Explainable to Explanatory Artificial Intelligence: Toward a New Paradigm for Human-Centered Explanations through Generative AI
• Automated Creation of the Legal Knowledge Graph Addressing Legislation on Violence Against Women: Resource, Methodology and Lessons Learned
• The Fair Game: Auditing & Debiasing AI Algorithms Over Time
• What Voting Rules Actually Do: A Data-Driven Analysis of Multi-Winner Voting
• Epidemic Control on a Large-Scale-Agent-Based Epidemiology Model using Deep Deterministic Policy Gradient
• SHACL Validation in the Presence of Ontologies: Semantics and Rewriting Techniques
• Automated Visualization Makeovers with LLMs
• Request-Only Optimization for Recommendation Systems
• Query-Aware Graph Neural Networks for Enhanced Retrieval-Augmented Generation
• AquiLLM: a RAG Tool for Capturing Tacit Knowledge in Research Groups
• OmniBench-RAG: A Multi-Domain Evaluation Platform for Retrieval-Augmented Generation Tools
• Lessons from A Large Language Model-based Outdoor Trail Recommendation Chatbot with Retrieval Augmented Generation
• Modeling Interactive Narrative Systems: A Formal Approach
• Comparison of Information Retrieval Techniques Applied to IT Support Tickets
• Beyond Single Labels: Improving Conversational Recommendation through LLM-Powered Data Augmentation
• Open-Source Agentic Hybrid RAG Framework for Scientific Literature Review
• Zero-Shot Retrieval for Scalable Visual Search in a Two-Sided Marketplace
• From Static to Dynamic: A Streaming RAG Approach to Real-time Knowledge Base
• Enhancing Retrieval-Augmented Generation for Electric Power Industry Customer Support
• HySemRAG: A Hybrid Semantic Retrieval-Augmented Generation Framework for Automated Literature Synthesis and Methodological Gap Analysis
• ITDR: An Instruction Tuning Dataset for Enhancing Large Language Models in Recommendations
• A Survey of LLM-based Deep Search Agents: Paradigm, Optimization, Evaluation, and Challenges
• Fine-Tuning Vision-Language Models for Markdown Conversion of Financial Tables in Malaysian Audited Financial Reports
• Can LLMs effectively provide game-theoretic-based scenarios for cybersecurity?
• LMAR: Language Model Augmented Retriever for Domain-specific Knowledge Indexing
• Breaking the Top-$K$ Barrier: Advancing Top-$K$ Ranking Metrics Optimization in Recommender Systems
• Towards Effective Offensive Security LLM Agents: Hyperparameter Tuning, LLM as a Judge, and a Lightweight CTF Benchmark
• Principle-Guided Verilog Optimization: IP-Safe Knowledge Transfer via Local-Cloud Collaboration
• Adversarial Attacks on Reinforcement Learning-based Medical Questionnaire Systems: Input-level Perturbation Strategies and Medical Constraint Validation
• Are All Genders Equal in the Eyes of Algorithms? -- Analysing Search and Retrieval Algorithms for Algorithmic Gender Fairness
• Selection-Based Vulnerabilities: Clean-Label Backdoor Attacks in Active Learning
• Risk Analysis Techniques for Governed LLM-based Multi-Agent Systems
• Empirical Evaluation of AI-Assisted Software Package Selection: A Knowledge Graph Approach
• DMFI: Dual-Modality Fine-Tuning and Inference Framework for LLM-Based Insider Threat Detection
• Log2Sig: Frequency-Aware Insider Threat Detection via Multivariate Behavioral Signal Decomposition
• Multi-Faceted Large Embedding Tables for Pinterest Ads Ranking
• Semantic Reasoning Meets Numerical Precision: An LLM-Powered Multi-Agent System for Power Grid Control
• A Physiologically-Constrained Neural Network Digital Twin Framework for Replicating Glucose Dynamics in Type 1 Diabetes
• Klear-CodeTest: Scalable Test Case Generation for Code Reinforcement Learning
• CLAPP: The CLASS LLM Agent for Pair Programming
• UnGuide: Learning to Forget with LoRA-Guided Diffusion Models
• Few-Shot Deployment of Pretrained MRI Transformers in Brain Imaging Tasks
• From Imperfect Signals to Trustworthy Structure: Confidence-Aware Inference from Heterogeneous and Reliability-Varying Utility Data
• AI-Guided Exploration of Large-Scale Codebases
• Integrating Vision Foundation Models with Reinforcement Learning for Enhanced Object Interaction
• Towards Transparent Ethical AI: A Roadmap for Trustworthy Robotic Systems
• Do Machines Think Emotionally? Cognitive Appraisal Analysis of Large Language Models
• Do Ethical AI Principles Matter to Users? A Large-Scale Analysis of User Sentiment and Satisfaction
• Enhancing Software Vulnerability Detection Through Adaptive Test Input Generation Using Genetic Algorithm
• REFS: Robust EEG feature selection with missing multi-dimensional annotation for emotion recognition
• ASLSL: Adaptive shared latent structure learning with incomplete multi-modal physiological data for multi-dimensional emotional feature selection
• Prosocial Behavior Detection in Player Game Chat: From Aligning Human-AI Definitions to Efficient Annotation at Scale
• A 3DGS-Diffusion Self-Supervised Framework for Normal Estimation from a Single Image
• Bifrost-1: Bridging Multimodal LLMs and Diffusion Models with Patch-level CLIP Latents
• Multi-Armed Bandits-Based Optimization of Decision Trees
• Mildly Conservative Regularized Evaluation for Offline Reinforcement Learning
• Impact-driven Context Filtering For Cross-file Code Completion
• DAFMSVC: One-Shot Singing Voice Conversion with Dual Attention Mechanism and Flow Matching
• Learning by Teaching: Engaging Students as Instructors of Large Language Models in Computer Science Education
• ETA: Energy-based Test-time Adaptation for Depth Completion
• ECMF: Enhanced Cross-Modal Fusion for Multimodal Emotion Recognition in MER-SEMI Challenge
• Hand by Hand: LLM Driving EMS Assistant for Operational Skill Learning
• Crisp Attention: Regularizing Transformers via Structured Sparsity
• Improved Sub-Visible Particle Classification in Flow Imaging Microscopy via Generative AI-Based Image Synthesis
• Temporal Self-Rewarding Language Models: Decoupling Chosen-Rejected via Past-Future
• Adaptive Heterogeneous Graph Neural Networks: Bridging Heterophily and Heterogeneity
• Fourier-VLM: Compressing Vision Tokens in the Frequency Domain for Large Vision-Language Models
• DP-LLM: Runtime Model Adaptation with Dynamic Layer-wise Precision Assignment
• EvolvR: Self-Evolving Pairwise Reasoning for Story Evaluation to Enhance Generation
• ThematicPlane: Bridging Tacit User Intent and Latent Spaces for Image Generation
• Architecture-Aware Generalization Bounds for Temporal Networks: Theory and Fair Comparison Methodology
• Can Large Models Fool the Eye? A New Turing Test for Biological Animation
• Towards MR-Based Trochleoplasty Planning
• Bounding Distributional Shifts in World Modeling through Novelty Detection
• MeanAudio: Fast and Faithful Text-to-Audio Generation with Mean Flows
• Mask & Match: Learning to Recognize Handwritten Math with Self-Supervised Attention
• GCHR : Goal-Conditioned Hindsight Regularization for Sample-Efficient Reinforcement Learning
• FMCE-Net++: Feature Map Convergence Evaluation and Training
• LLM Serving Optimization with Variable Prefill and Decode Lengths
• Less is More: Selective Reflection for Compatible and Efficient Knowledge Distillation in Large Language Models
• Roll Your Eyes: Gaze Redirection via Explicit 3D Eyeball Rotation
• Semantic Item Graph Enhancement for Multimodal Recommendation
• One Size Does Not Fit All: A Distribution-Aware Sparsification for More Precise Model Merging
• UR$^2$: Unify RAG and Reasoning through Reinforcement Learning
• UW-3DGS: Underwater 3D Reconstruction with Physics-Aware Gaussian Splatting
• Synthetic Data-Driven Multi-Architecture Framework for Automated Polyp Segmentation Through Integrated Detection and Mask Generation
• Differentially Private Federated Clustering with Random Rebalancing
• Benchmarking Pretrained Molecular Embedding Models For Molecular Representation Learning
• LoRA in LoRA: Towards Parameter-Efficient Architecture Expansion for Continual Visual Instruction Tuning
• Classification is a RAG problem: A case study on hate speech detection