AI Research News Feeds for October 20th, 2025

AI RESEARCH PAPERS & ACADEMIC SOURCES

• CuSfM: CUDA-Accelerated Structure-from-Motion
• QCFace: Image Quality Control for boosting Face Representation & Recognition
• Proto-Former: Unified Facial Landmark Detection by Prototype Transformer
• SHARE: Scene-Human Aligned Reconstruction
• Adaptive transfer learning for surgical tool presence detection in laparoscopic videos through gradual freezing fine-tuning
• FreqPDE: Rethinking Positional Depth Embedding for Multi-View 3D Object Detection Transformers
• PFGS: Pose-Fused 3D Gaussian Splatting for Complete Multi-Pose Object Reconstruction
• Rethinking Convergence in Deep Learning: The Predictive-Corrective Paradigm for Anatomy-Informed Brain MRI Segmentation
• MAVR-Net: Robust Multi-View Learning for MAV Action Recognition with Cross-View Attention
• DPTrack:Directional Kernel-Guided Prompt Learning for Robust Nighttime Aerial Tracking
• Improving Micro-Expression Recognition with Phase-Aware Temporal Augmentation
• MRASfM: Multi-Camera Reconstruction and Aggregation through Structure-from-Motion in Driving Scenes
• MSAM: Multi-Semantic Adaptive Mining for Cross-Modal Drone Video-Text Retrieval
• A Novel Combined Optical Flow Approach for Comprehensive Micro-Expression Recognition
• Iterative Motion Compensation for Canonical 3D Reconstruction from UAV Plant Images Captured in Windy Conditions
• Rethinking Efficient Hierarchical Mixing Architecture for Low-light RAW Image Enhancement
• Exploring Conditions for Diffusion models in Robotic Control
• Balanced Multi-Task Attention for Satellite Image Classification: A Systematic Approach to Achieving 97.23% Accuracy on EuroSAT Without Pre-Training
• Diffusion Bridge Networks Simulate Clinical-grade PET from MRI for Dementia Diagnostics
• Imaginarium: Vision-guided High-Quality 3D Scene Layout Generation
• Unmasking Facial DeepFakes: A Robust Multiview Detection Framework for Natural Images
• Standardization for improved Spatio-Temporal Image Fusion
• FlexiReID: Adaptive Mixture of Expert for Multi-Modal Person Re-Identification
• Quantized FCA: Efficient Zero-Shot Texture Anomaly Detection
• Lightweight Data-Free Denoising for Detail-Preserving Biomedical Image Restoration
• Deep Learning Based Domain Adaptation Methods in Remote Sensing: A Comprehensive Survey
• Uncertainty-Aware Extreme Point Tracing for Weakly Supervised Ultrasound Image Segmentation
• Unimedvl: Unifying Medical Multimodal Understanding And Generation Through Observation-Knowledge-Analysis
• Scaling Instruction-Based Video Editing with a High-Quality Synthetic Dataset
• SEGA: A Stepwise Evolution Paradigm for Content-Aware Layout Generation with Design Prior
• ReCon: Region-Controllable Data Augmentation with Rectification and Alignment for Object Detection
• ERNet: Efficient Non-Rigid Registration Network for Point Sequences
• VISTA: A Test-Time Self-Improving Video Generation Agent
• Neuro-Symbolic Spatial Reasoning in Segmentation
• 3DPR: Single Image 3D Portrait Relight using Generative Priors
• Memory-SAM: Human-Prompt-Free Tongue Segmentation via Retrieval-to-Prompt
• BLIP3o-NEXT: Next Frontier of Native Image Generation
• BiomedXPro: Prompt Optimization for Explainable Diagnosis with Biomedical Vision Language Models
• LightsOut: Diffusion-based Outpainting for Enhanced Lens Flare Removal
• Skyfall-GS: Synthesizing Immersive 3D Urban Scenes from Satellite Imagery
• Neural Posterior Estimation for Cataloging Astronomical Images from the Legacy Survey of Space and Time
• Confidence-Weighted Semi-Supervised Learning for Skin Lesion Segmentation Using Hybrid CNN-Transformer Networks
• Fix False Transparency by Noise Guided Splatting
• SANR: Scene-Aware Neural Representation for Light Field Image Compression with Rate-Distortion Optimization
• CAIT: Triple-Win Compression towards High Accuracy, Fast Inference, and Favorable Transferability For ViTs
• Diffusion Models are Efficient Data Generators for Human Mesh Recovery
• CLOVER: Context-aware Long-term Object Viewpoint- and Environment- Invariant Representation Learning
• Self-supervised Multi-future Occupancy Forecasting for Autonomous Driving
• V2X-Radar: A Multi-modal Dataset with 4D Radar for Cooperative Perception
• PrefixKV: Adaptive Prefix KV Cache is What Vision Instruction-Following Models Need for Efficient Generation
• Elevating Visual Perception in Multimodal LLMs with Visual Embedding Distillation
• Conformal Risk Control for Pulmonary Nodule Detection
• Unleashing the Potential of Pre-Trained Diffusion Models for Generalizable Person Re-Identification
• YOLOE: Real-Time Seeing Anything
• L2RSI: Cross-view LiDAR-based Place Recognition for Large-scale Urban Scenes via Remote Sensing Imagery
• UniMamba: Unified Spatial-Channel Representation Learning with Group-Efficient Mamba for LiDAR-based 3D Object Detection
• A Plug-and-Play Learning-based IMU Bias Factor for Robust Visual-Inertial Odometry
• Bolt3D: Generating 3D Scenes in Seconds
• CHROME: Clothed Human Reconstruction with Occlusion-Resilience and Multiview-Consistency from a Single Image
• X$^{2}$-Gaussian: 4D Radiative Gaussian Splatting for Continuous-time Tomographic Reconstruction
• CMaP-SAM: Contraction Mapping Prior for SAM-driven Few-shot Segmentation
• CADE 2.5 - ZeResFDG: Frequency-Decoupled, Rescaled and Zero-Projected Guidance for SD/SDXL Latent Diffusion Models
• Scope: Selective Cross-modal Orchestration of Visual Perception Experts
• CoDS: Enhancing Collaborative Perception in Heterogeneous Scenarios via Domain Separation
• LOPR: Latent Occupancy PRediction using Generative Models
• TAS: A Transit-Aware Strategy for Embodied Navigation with Non-Stationary Targets
• Universal Vessel Segmentation for Multi-Modality Retinal Images
• MLFM: Multi-Layered Feature Maps for Richer Language Understanding in Zero-Shot Semantic Navigation
• A Generalizable Rhetorical Strategy Annotation Model Using LLM-based Debate Simulation and Labelling
• Measuring the Effect of Disfluency in Multilingual Knowledge Probing Benchmarks
• Scaling Beyond Context: A Survey of Multimodal Retrieval-Augmented Generation for Document Understanding
• Automatic essay scoring: leveraging Jaccard coefficient and Cosine similaritywith n-gram variation in vector space model approach
• Accelerating Mobile Language Model Generation via Hybrid Context and Hardware Coordination
• Capabilities and Evaluation Biases of Large Language Models in Classical Chinese Poetry Generation: A Case Study on Tang Poetry
• AutoGraph-R1: End-to-End Reinforcement Learning for Knowledge Graph Construction
• Infinity Parser: Layout Aware Reinforcement Learning for Scanned Document Parsing
• VocalBench-DF: A Benchmark for Evaluating Speech LLM Robustness to Disfluency
• Large-scale User Game Lifecycle Representation Learning
• When Seeing Is not Enough: Revealing the Limits of Active Reasoning in MLLMs
• Controllable Abstraction in Summary Generation for Large Language Models via Prompt Engineering
• CORE: Reducing UI Exposure in Mobile Agents via Collaboration Between Cloud and Local LLMs
• Temporal Referential Consistency: Do LLMs Favor Sequences Over Absolute Time References?
• From Characters to Tokens: Dynamic Grouping with Hierarchical BPE
• Latent Reasoning in LLMs as a Vocabulary-Space Superposition
• Finetuning LLMs for EvaCun 2025 token prediction shared task
• From Ghazals to Sonnets: Decoding the Polysemous Expressions of Love Across Languages
• BiMax: Bidirectional MaxSim Score for Document-Level Alignment
• The Elephant in the Coreference Room: Resolving Coreference in Full-Length French Fiction Works
• HypoSpace: Evaluating LLM Creativity as Set-Valued Hypothesis Generators under Underdetermination
• Leveraging LLMs for Context-Aware Implicit Textual and Multimodal Hate Speech Detection
• Cost-Aware Retrieval-Augmentation Reasoning Models with Adaptive Retrieval Depth
• Emergence of Linear Truth Encodings in Language Models
• Paper2Web: Let's Make Your Paper Alive!
• Shakti-VLMs: Scalable Vision-Language Models for Enterprise AI
• Train a Unified Multimodal Data Quality Classifier with Synthetic Data
• MAGPIE: A benchmark for Multi-AGent contextual PrIvacy Evaluation
• Leveraging Test Driven Development with Large Language Models for Reliable and Verifiable Spreadsheet Code Generation: A Research Framework
• SQuAI: Scientific Question-Answering with Multi-Agent Retrieval-Augmented Generation
• GraphMind: Interactive Novelty Assessment System for Accelerating Scientific Discovery
• Evaluating Large Language Models with Psychometrics
• Cross-layer Attention Sharing for Pre-trained Large Language Models
• To Err Is Human; To Annotate, SILICON? Reducing Measurement Error in LLM Annotation
• Event Segmentation Applications in Large Language Model Enabled Automated Recall Assessments
• Towards Human Cognition: Visual Context Guides Syntactic Priming in Fusion-Encoded Models
• Generating patient cohorts from electronic health records using two-step retrieval-augmented text-to-SQL generation
• Summarizing Speech: A Comprehensive Survey
• Deliberation on Priors: Trustworthy Reasoning of Large Language Models on Knowledge Graphs
• RAGRouter: Learning to Route Queries to Multiple Retrieval-Augmented Language Models
• PBEBench: A Multi-Step Programming by Examples Reasoning Benchmark inspired by Historical Linguistics
• What's Wrong with Your Code Generated by Large Language Models? An Extensive Study
• MemeSense: An Adaptive In-Context Framework for Social Commonsense Driven Meme Moderation
• Toward Safe and Human-Aligned Game Conversational Recommendation via Multi-Agent Decomposition
• NANO3D: A Training-Free Approach for Efficient 3D Editing Without Masks
• Constantly Improving Image Models Need Constantly Improving Benchmarks
• LoRAverse: A Submodular Framework to Retrieve Diverse Adapters for Diffusion Models
• MOBIUS: Big-to-Mobile Universal Instance Segmentation via Multi-modal Bottleneck Fusion and Calibrated Decoder Pruning
• Generalized Dynamics Generation towards Scannable Physical World Model
• Directional Reasoning Injection for Fine-Tuning MLLMs
• A solution to generalized learning from small training sets found in everyday infant experiences
• SaLon3R: Structure-aware Long-term Generalizable 3D Reconstruction from Unposed Images
• TGT: Text-Grounded Trajectories for Locally Controlled Video Generation
• Fourier Transform Multiple Instance Learning for Whole Slide Image Classification
• Hyperparameter Optimization and Reproducibility in Deep Learning Model Training
• Salient Concept-Aware Generative Data Augmentation
• CARDIUM: Congenital Anomaly Recognition with Diagnostic Images and Unified Medical records
• The Face of Persuasion: Analyzing Bias and Generating Culture-Aware Ads
• DriveGen3D: Boosting Feed-Forward Driving Scene Generation with Efficient Video Diffusion
• RadioDiff-$k^2$: Helmholtz Equation Informed Generative Diffusion Model for Multi-Path Aware Radio Map Construction
• Euclidean Distance Matrix Completion via Asymmetric Projected Gradient Descent
• MegaScale-MoE: Large-Scale Communication-Efficient Training of Mixture-of-Experts Models in Production
• msf-CNN: Patch-based Multi-Stage Fusion with Convolutional Neural Networks for TinyML
• Conditional Generative Modeling for Enhanced Credit Risk Management in Supply Chain Finance
• Understanding Generalization in Node and Link Prediction
• A Weakly Supervised Transformer for Rare Disease Diagnosis and Subphenotyping from EHRs with Pulmonary Case Studies
• A Cycle-Consistency Constrained Framework for Dynamic Solution Space Reduction in Noninjective Regression
• LeMat-Traj: A Scalable and Unified Dataset of Materials Trajectories for Atomistic Modeling
• Learning Unified Representations from Heterogeneous Data for Robust Heart Rate Modeling
• Traces Propagation: Memory-Efficient and Scalable Forward-Only Learning in Spiking Neural Networks
• Machine Learning-Based Ultrasonic Weld Characterization Using Hierarchical Wave Modeling and Diffusion-Driven Distribution Alignment
• When In Doubt, Abstain: The Impact of Abstention on Strategic Classification
• How Sparse Can We Prune A Deep Network: A Fundamental Limit Perspective
• Which exceptional low-dimensional projections of a Gaussian point cloud can be found in polynomial time?
• Spatial Supply Repositioning with Censored Demand Data
• End-to-End Learning Framework for Solving Non-Markovian Optimal Control
• DeepRV: Accelerating spatiotemporal inference with pre-trained neural priors
• Landmark-Based Node Representations for Shortest Path Distance Approximations in Random Graphs
• SPICE: A Synergistic, Precise, Iterative, and Customizable Image Editing Workflow
• Low-Rank Adaptation of Neural Fields
• SYMI: Efficient Mixture-of-Experts Training via Model and Optimizer State Decoupling
• Improving Inference-Time Optimisation for Vocal Effects Style Transfer with a Gaussian Prior
• Explainable Machine Learning for Oxygen Diffusion in Perovskites and Pyrochlores
• Uncertainty Quantification for Physics-Informed Neural Networks with Extended Fiducial Inference
• VLMLight: Safety-Critical Traffic Signal Control via Vision-Language Meta-Control and Dual-Branch Reasoning Architecture
• Onboard Mission Replanning for Adaptive Cooperative Multi-Robot Systems
• Implicit neural representations for accurate estimation of the standard model of white matter
• Autonomous Cyber Resilience via a Co-Evolutionary Arms Race within a Fortified Digital Twin Sandbox
• Operationalizing Automated Essay Scoring: A Human-Aware Approach
• Mind the Gap: Navigating Inference with Optimal Transport Maps
• Meta-learning of Gibbs states for many-body Hamiltonians with applications to Quantum Boltzmann Machines
• Migration as a Probe: A Generalizable Benchmark Framework for Specialist vs. Generalist Machine-Learned Force Fields
• KGAlign: Joint Semantic-Structural Knowledge Encoding for Multimodal Fake News Detection
• UNet with Self-Adaptive Mamba-Like Attention and Causal-Resonance Learning for Medical Image Segmentation
• DiEmo-TTS: Disentangled Emotion Representations via Self-Supervised Distillation for Cross-Speaker Emotion Transfer in Text-to-Speech
• EmoSphere-SER: Enhancing Speech Emotion Recognition Through Spherical Representation with Auxiliary Classification
• Topological Structure Learning Should Be A Research Priority for LLM-Based Multi-Agent Systems
• Hyperbolic Dataset Distillation
• FinChain: A Symbolic Benchmark for Verifiable Chain-of-Thought Financial Reasoning
• When Does Closeness in Distribution Imply Representational Similarity? An Identifiability Perspective
• Refer to Any Segmentation Mask Group With Vision-Language Prompts
• MOBODY: Model Based Off-Dynamics Offline Reinforcement Learning
• FinHEAR: Human Expertise and Adaptive Risk-Aware Temporal Reasoning for Financial Decision-Making
• General-Purpose Robotic Navigation via LVLM-Orchestrated Perception, Reasoning, and Acting
• Clarifying the Ti-V Phase Diagram Using First-Principles Calculations and Bayesian Learning
• Turning Sand to Gold: Recycling Data to Bridge On-Policy and Off-Policy Learning via Causal Bound
• Extending Load Forecasting from Zonal Aggregates to Individual Nodes for Transmission System Operators
• ES-C51: Expected Sarsa Based C51 Distributional Reinforcement Learning Algorithm
• AlignFlow: Improving Flow-based Generative Models with Semi-Discrete Optimal Transport
• IQNN-CS: Interpretable Quantum Neural Network for Credit Scoring
• Internalizing World Models via Self-Play Finetuning for Agentic RL
• Learn to Change the World: Multi-level Reinforcement Learning with Model-Changing Actions
• Antislop: A Comprehensive Framework for Identifying and Eliminating Repetitive Patterns in Language Models
• Physics-informed data-driven machine health monitoring for two-photon lithography
• Online Correlation Clustering: Simultaneously Optimizing All $\ell_p$-norms
• Navigating the consequences of mechanical ventilation in clinical intensive care settings through an evolutionary game-theoretic framework
• A Simple Method for PMF Estimation on Large Supports
• Predicting the Unpredictable: Reproducible BiLSTM Forecasting of Incident Counts in the Global Terrorism Database (GTD)
• Policy Transfer Ensures Fast Learning for Continuous-Time LQR with Entropy Regularization
• A simple mean field model of feature learning
• Finding geodesics with the Deep Ritz method
• An Advanced Two-Stage Model with High Sensitivity and Generalizability for Prediction of Hip Fracture Risk Using Multiple Datasets
• Dissecting Mahalanobis: How Feature Geometry and Normalization Shape OOD Detection
• Soundness-Aware Level: A Microscopic Signature that Predicts LLM Reasoning Potential
• Reflections from Research Roundtables at the Conference on Health, Inference, and Learning (CHIL) 2025
• Machine Learning for Early Detection of Meningitis: Stacked Ensemble Learning with EHR data
• Integrating Product Coefficients for Improved 3D LiDAR Data Classification (Part II)
• Stress-Aware Learning under KL Drift via Trust-Decayed Mirror Descent
• FinTrust: A Comprehensive Benchmark of Trustworthiness Evaluation in Finance Domain
• Dual-Weighted Reinforcement Learning for Generative Preference Modeling
• Spatiotemporal Transformers for Predicting Avian Disease Risk from Migration Trajectories
• Causal Time Series Modeling of Supraglacial Lake Evolution in Greenland under Distribution Shift
• Semi-Supervised Regression with Heteroscedastic Pseudo-Labels
• Small Ensemble-based Data Assimilation: A Machine Learning-Enhanced Data Assimilation Method with Limited Ensemble Size
• DFCA: Decentralized Federated Clustering Algorithm
• On the Generalization Properties of Learning the Random Feature Models with Learnable Activation Functions
• Backdoor or Manipulation? Graph Mixture of Experts Can Defend Against Various Graph Adversarial Attacks
• Sequence Modeling with Spectral Mean Flows
• Iterative Refinement of Flow Policies in Probability Space for Online Reinforcement Learning
• Geometric Mixture Models for Electrolyte Conductivity Prediction
• Online Kernel Dynamic Mode Decomposition for Streaming Time Series Forecasting with Adaptive Windowing
• ParaFormer: Shallow Parallel Transformers with Progressive Approximation
• Safe, Efficient, and Robust Reinforcement Learning for Ranking and Diffusion Models
• Particle Dynamics for Latent-Variable Energy-Based Models
• Adversary-Free Counterfactual Prediction via Information-Regularized Representations
• Theoretical Refinement of CLIP by Utilizing Linear Structure of Optimal Similarity
• Compressive Modeling and Visualization of Multivariate Scientific Data using Implicit Neural Representation
• An Empirical Study on MC Dropout--Based Uncertainty--Error Correlation in 2D Brain Tumor Segmentation
• Doubly Robust Estimation of Causal Effects in Strategic Equilibrium Systems
• On the Neural Feature Ansatz for Deep Neural Networks
• Attn-JGNN: Attention Enhanced Join-Graph Neural Networks
• GRATING: Low-Latency and Memory-Efficient Semantic Selection on Device
• Decentralized Parameter-Free Online Learning
• Deep Neural ODE Operator Networks for PDEs
• Fast and Compact Tsetlin Machine Inference on CPUs Using Instruction-Level Optimization
• WARP-LUTs - Walsh-Assisted Relaxation for Probabilistic Look Up Tables
• Constrained Adversarial Perturbation
• A Comprehensive Evaluation of Graph Neural Networks and Physics Informed Learning for Surrogate Modelling of Finite Element Analysis
• SAMix: Calibrated and Accurate Continual Learning via Sphere-Adaptive Mixup and Neural Collapse
• Poultry Farm Intelligence: An Integrated Multi-Sensor AI Platform for Enhanced Welfare and Productivity
• Cavity Duplexer Tuning with 1d Resnet-like Neural Networks
• FIDDLE: Reinforcement Learning for Quantum Fidelity Enhancement
• Transfer Orthology Networks
• Learning Correlated Reward Models: Statistical Barriers and Opportunities
• FIRE: Fact-checking with Iterative Retrieval and Verification
• Estimand framework and intercurrent events handling for clinical trials with time-to-event outcomes
• Reliable data clustering with Bayesian community detection
• The Tree-SNE Tree Exists
• Composition-Grounded Instruction Synthesis for Visual Reasoning
• Comprehensive language-image pre-training for 3D medical image understanding
• The Minimax Lower Bound of Kernel Stein Discrepancy Estimation
• PoTS: Proof-of-Training-Steps for Backdoor Detection in Large Language Models
• Polarization based direction of arrival estimation using a radio interferometric array
• Deep generative priors for 3D brain analysis
• Beyond PCA: Manifold Dimension Estimation via Local Graph Structure
• OCR-APT: Reconstructing APT Stories from Audit Logs using Subgraph Anomaly Detection and LLMs
• HyperAIRI: a plug-and-play algorithm for precise hyperspectral image reconstruction in radio interferometry
• How to Sell High-Dimensional Data Optimally
• HOB: A Holistically Optimized Bidding Strategy under Heterogeneous Auction Mechanisms with Organic Traffic
• Minimisation of Submodular Functions Using Gaussian Zeroth-Order Random Oracles
• Foresighted Online Policy Optimization with Interference
• Hyperbolic Structured Classification for Robust Single Positive Multi-label Learning
• Layer as Puzzle Pieces: Compressing Large Language Models through Layer Concatenation
• Transfer Learning for Benign Overfitting in High-Dimensional Linear Regression
• Singularity-free dynamical invariants-based quantum control
• RankSEG-RMA: An Efficient Segmentation Algorithm via Reciprocal Moment Approximation
• TranSimHub:A Unified Air-Ground Simulation Platform for Multi-Modal Perception and Decision-Making
• Recursive Inference for Heterogeneous Multi-Output GP State-Space Models with Arbitrary Moment Matching
• LILAC: Long-sequence Incremental Low-latency Arbitrary Motion Stylization via Streaming VAE-Diffusion with Causal Decoding
• Information Theory in Open-world Machine Learning Foundations, Frameworks, and Future Direction
• Semantic4Safety: Causal Insights from Zero-shot Street View Imagery Segmentation for Urban Road Safety
• Nonlinear Dimensionality Reduction Techniques for Bayesian Optimization
• Online Policy Learning via a Self-Normalized Maximal Inequality
• AI and analytics in sports: Leveraging BERTopic to map the past and chart the future
• Latent Feature Alignment: Discovering Biased and Interpretable Subpopulations in Face Recognition Models
• VO-DP: Semantic-Geometric Adaptive Diffusion Policy for Vision-Only Robotic Manipulation
• SpikeFit: Towards Optimal Deployment of Spiking Networks on Neuromorphic Hardware
• Geometric Convergence Analysis of Variational Inference via Bregman Divergences
• Kernel-Based Evaluation of Conditional Biological Sequence Models
• Stochastic Optimization with Random Search
• GOGH: Correlation-Guided Orchestration of GPUs in Heterogeneous Clusters
• Bayesian Inference for PDE-based Inverse Problems using the Optimization of a Discrete Loss
• Disentanglement of Sources in a Multi-Stream Variational Autoencoder
• A Split-Client Approach to Second-Order Optimization
• QSilk: Micrograin Stabilization and Adaptive Quantile Clipping for Detail-Friendly Latent Diffusion
• On Non-interactive Evaluation of Animal Communication Translators
• Towards more holistic interpretability: A lightweight disentangled Concept Bottleneck Model
• Enhanced Renewable Energy Forecasting using Context-Aware Conformal Prediction
• DexCanvas: Bridging Human Demonstrations and Robot Learning for Dexterous Manipulation
• On Universality of Deep Equivariant Networks
• Error analysis of a compositional score-based algorithm for simulation-based inference
• Blackwell's Approachability for Sequential Conformal Inference
• SpeechLLMs for Large-scale Contextualized Zero-shot Slot Filling
• Personalized Semi-Supervised Federated Learning for Human Activity Recognition
• Photovoltaic power forecasting using quantum machine learning
• Optimistic Query Routing in Clustering-based Approximate Maximum Inner Product Search
• METIS: Fast Quality-Aware RAG Systems with Configuration Adaptation
• A Multimodal Lightweight Approach to Fault Diagnosis of Induction Motors in High-Dimensional Dataset
• Scalable Multi-phase Word Embedding Using Conjunctive Propositional Clauses
• Privacy-Preserving Dataset Combination
• Closed-Form Training Dynamics Reveal Learned Features and Linear Structure in Word2Vec-like Models
• Predicting gene essentiality and drug response from perturbation screens in preclinical cancer models with LEAP: Layered Ensemble of Autoencoders and Predictors
• All Roads Lead to Likelihood: The Value of Reinforcement Learning in Fine-Tuning
• Rethinking Robustness in Machine Learning: A Posterior Agreement Approach
• LayerCraft: Enhancing Text-to-Image Generation with CoT Reasoning and Layered Object Integration
• Neural Mean-Field Games: Extending Mean-Field Game Theory with Neural Stochastic Differential Equations
• PokeeResearch: Effective Deep Research via Reinforcement Learning from AI Feedback and Robust Reasoning Scaffold
• Spatial457: A Diagnostic Benchmark for 6D Spatial Reasoning of Large Multimodal Models
• End-to-End Multi-Modal Diffusion Mamba
• Design and Analysis of Parallel Artificial Protozoa Optimizer (P-APO) using CUDA Architecture
• DeepAries: Adaptive Rebalancing Interval Selection for Enhanced Portfolio Selection
• RegimeFolio: A Regime Aware ML System for Sectoral Portfolio Optimization in Dynamic Markets
• Constrained Diffusion for Protein Design with Hard Structural Constraints
• The Role of Federated Learning in Improving Financial Security: A Survey
• GAZE:Governance-Aware pre-annotation for Zero-shot World Model Environments
• PC-UNet: An Enforcing Poisson Statistics U-Net for Positron Emission Tomography Denoising
• Evaluation and Implementation of Machine Learning Algorithms to Predict Early Detection of Kidney and Heart Disease in Diabetic Patients
• VaultGemma: A Differentially Private Gemma Model
• Automated Snippet-Alignment Data Augmentation for Code Translation
• TangledFeatures: Robust Feature Selection in Highly Correlated Spaces
• Rethinking Toxicity Evaluation in Large Language Models: A Multi-Label Perspective
• Can generative AI figure out figurative language? The influence of idioms on essay scoring by ChatGPT, Gemini, and Deepseek
• Hybrid Autoencoder-Based Framework for Early Fault Detection in Wind Turbines
• From Universal Approximation Theorem to Tropical Geometry of Multi-Layer Perceptrons
• DeLeaker: Dynamic Inference-Time Reweighting For Semantic Leakage Mitigation in Text-to-Image Models
• Active Honeypot Guardrail System: Probing and Confirming Multi-Turn LLM Jailbreaks
• UrbanVerse: Scaling Urban Simulation by Watching City-Tour Videos
• The Coverage Principle: How Pre-training Enables Post-Training
• Sequential Comics for Jailbreaking Multimodal Large Language Models via Structured Visual Storytelling
• DMRetriever: A Family of Models for Improved Text Retrieval in Disaster Management
• Beyond Outcome-Based Imperfect-Recall: Higher-Resolution Abstractions for Imperfect-Information Games
• Operator Flow Matching for Timeseries Forecasting
• Continual Learning via Sparse Memory Finetuning
• Targeted Attacks and Defenses for Distributed Federated Learning in Vehicular Networks
• DLER: Doing Length pEnalty Right - Incentivizing More Intelligence per Token via Reinforcement Learning
• Latent Topic Synthesis: Leveraging LLMs for Electoral Ad Analysis
• FarsiMCQGen: a Persian Multiple-choice Question Generation Framework
• XModBench: Benchmarking Cross-Modal Capabilities and Consistency in Omni-Language Models
• Structure-R1: Dynamically Leveraging Structural Knowledge in LLM Reasoning through Reinforcement Learning
• The Economics of AI Foundation Models: Openness, Competition, and Governance
• Automotive Crash Dynamics Modeling Accelerated with Machine Learning
• ReasonIF: Large Reasoning Models Fail to Follow Instructions During Reasoning
• Extending Audio Context for Long-Form Understanding in Large Audio-Language Models
• Adaptive Individual Uncertainty under Out-Of-Distribution Shift with Expert-Routed Conformal Prediction
• Planner and Executor: Collaboration between Discrete Diffusion And Autoregressive Models in Reasoning
• DRO-InstructZero: Distributionally Robust Prompt Optimization for Large Language Models
• Robust Layerwise Scaling Rules by Proper Weight Decay Tuning
• TraceCoder: Towards Traceable ICD Coding via Multi-Source Knowledge Integration
• TACL: Threshold-Adaptive Curriculum Learning Strategy for Enhancing Medical Text Understanding
• Foundation Models for Scientific Discovery: From Paradigm Enhancement to Paradigm Transition
• Post-Processing Methods for Improving Accuracy in MRI Inpainting
• Exemplar-Guided Planing: Enhanced LLM Agent for KGQA
• MTmixAtt: Integrating Mixture-of-Experts with Multi-Mix Attention for Large-Scale Recommendation
• Identifying internal patterns in (1+1)-dimensional directed percolation using neural networks
• VERA-MH Concept Paper
• Latent Diffusion Model without Variational Autoencoder
• DSSmoothing: Toward Certified Dataset Ownership Verification for Pre-trained Language Models via Dual-Space Smoothing
• BeLLMan: Controlling LLM Congestion
• ASBI: Leveraging Informative Real-World Data for Active Black-Box Simulator Tuning
• Readability Reconsidered: A Cross-Dataset Analysis of Reference-Free Metrics
• When to Ensemble: Identifying Token-Level Points for Stable and Fast LLM Ensembling
• GaussGym: An open-source real-to-sim framework for learning locomotion from pixels
• Kernel Regression in Structured Non-IID Settings: Theory and Implications for Denoising Score Learning
• Cortical-SSM: A Deep State Space Model for EEG and ECoG Motor Imagery Decoding
• Towards Robust Zero-Shot Reinforcement Learning
• DroneAudioset: An Audio Dataset for Drone-based Search and Rescue
• MARIS: Marine Open-Vocabulary Instance Segmentation with Geometric Enhancement and Semantic Alignment
• Robust High-Resolution Multi-Organ Diffusion MRI Using Synthetic-Data-Tuned Prompt Learning
• Fine-Tuning MedGemma for Clinical Captioning to Enhance Multimodal RAG over Malaysia CPGs
• Learning to Detect Unknown Jailbreak Attacks in Large Vision-Language Models
• Select Less, Reason More: Prioritizing Evidence Purity for Video Reasoning
• A Theoretical Study on Bridging Internal Probability and Self-Consistency for LLM Reasoning
• Expediting Reinforcement Learning by Incorporating Knowledge About Temporal Causality in the Environment
• Robust Optimization in Causal Models and G-Causal Normalizing Flows
• Learning to Answer from Correct Demonstrations
• SoK: Taxonomy and Evaluation of Prompt Security in Large Language Models
• Selecting and Combining Large Language Models for Scalable Code Clone Detection
• An Experimental Study of Real-Life LLM-Proposed Performance Improvements
• OffSim: Offline Simulator for Model-based Offline Inverse Reinforcement Learning
• DeceptionBench: A Comprehensive Benchmark for AI Deception Behaviors in Real-world Scenarios
• The Road Less Traveled: Enhancing Exploration in LLMs via Sequential Sampling
• AI Adoption in NGOs: A Systematic Literature Review
• Language Models are Injective and Hence Invertible
• Revisiting Knowledge Distillation: The Hidden Role of Dataset Size
• MCA: Modality Composition Awareness for Robust Composed Multimodal Retrieval
• TokenTiming: A Dynamic Alignment Method for Universal Speculative Decoding Model Pairs
• Rethinking Cross-lingual Gaps from a Statistical Viewpoint
• Think Parallax: Solving Multi-Hop Problems via Multi-View Knowledge-Graph-Based Retrieval-Augmented Generation
• ClapperText: A Benchmark for Text Recognition in Low-Resource Archival Documents
• KITE: A Benchmark for Evaluating Korean Instruction-Following Abilities in Large Language Models
• SpikeVox: Towards Energy-Efficient Speech Therapy Framework with Spike-driven Generative Language Models
• The Spark Effect: On Engineering Creative Diversity in Multi-Agent AI Systems
• Lightweight CycleGAN Models for Cross-Modality Image Transformation and Experimental Quality Assessment in Fluorescence Microscopy
• CQD-SHAP: Explainable Complex Query Answering via Shapley Values
• Enhance Large Language Models as Recommendation Systems with Collaborative Filtering
• Valeo Near-Field: a novel dataset for pedestrian intent detection
• CarBoN: Calibrated Best-of-N Sampling Improves Test-time Reasoning
• ProofBridge: Auto-Formalization of Natural Language Proofs in Lean via Joint Embeddings
• Mixture of Experts Approaches in Dense Retrieval Tasks
• Towards Label-Free Brain Tumor Segmentation: Unsupervised Learning with Multimodal MRI
• KS-Net: Multi-layer network model for determining the rotor type from motor parameters in interior PMSMs
• Exploring the Synergy of Quantitative Factors and Newsflow Representations from Large Language Models for Stock Return Prediction
• ProofOptimizer: Training Language Models to Simplify Proofs without Human Demonstrations
• Beyond-Diagonal RIS Under Non-Idealities: Learning-Based Architecture Discovery and Optimization
• ProSh: Probabilistic Shielding for Model-free Reinforcement Learning
• DGME-T: Directional Grid Motion Encoding for Transformer-Based Historical Camera Movement Classification
• RLAF: Reinforcement Learning from Automaton Feedback
• Attention Sinks in Diffusion Language Models
• LLMs Judge Themselves: A Game-Theoretic Framework for Human-Aligned Evaluation
• NDM: A Noise-driven Detection and Mitigation Framework against Implicit Sexual Intentions in Text-to-Image Generation
• Semantic segmentation with coarse annotations
• Controlling the image generation process with parametric activation functions
• AB-UPT for Automotive and Aerospace Applications
• Chronos-2: From Univariate to Universal Forecasting
• GENESIS: A Generative Model of Episodic-Semantic Interaction
• SNOO: Step-K Nesterov Outer Optimizer - The Surprising Effectiveness of Nesterov Momentum Applied to Pseudo-Gradients
• Enhanced Sentiment Interpretation via a Lexicon-Fuzzy-Transformer Framework
• Self-Certifying Primal-Dual Optimization Proxies for Large-Scale Batch Economic Dispatch
• InfiMed-ORBIT: Aligning LLMs on Open-Ended Complex Tasks via Rubric-Based Incremental Training
• PolySkill: Learning Generalizable Skills Through Polymorphic Abstraction
• OmniVinci: Enhancing Architecture and Data for Omni-Modal Understanding LLM
• Flexora: Flexible Low Rank Adaptation for Large Language Models
• Beyond Static Assumptions: the Predictive Justified Perspective Model for Epistemic Planning
• Where Common Knowledge Cannot Be Formed, Common Belief Can -- Planning with Multi-Agent Belief Using Group Justified Perspectives
• FERA: Foil Fencing Referee Assistant Using Pose-Based Multi-Label Move Recognition and Rule Reasoning
• Establishing trust in automated reasoning
• MotionScript: Natural Language Descriptions for Expressive 3D Human Motions
• BandCondiNet: Parallel Transformers-based Conditional Popular Music Generation with Multi-View Features
• Retrieval-Augmented Test Generation: How Far Are We?
• Memory-Efficient Large Language Models for Program Repair with Semantic-Guided Patch Generation
• Variational Autoencoders for Efficient Simulation-Based Inference
• Competition and Diversity in Generative AI
• Towards smart and adaptive agents for active sensing on edge devices
• Retro3D: A 3D-aware Template-free Method for Enhancing Retrosynthesis via Molecular Conformer Information
• GuardReasoner: Towards Reasoning-based LLM Safeguards
• FEMBA: Efficient and Scalable EEG Analysis with a Bidirectional Mamba Foundation Model
• Methods and Trends in Detecting AI-Generated Images: A Comprehensive Review
• Finetuning and Quantization of EEG-Based Foundational BioSignal Models on ECG and PPG Data for Blood Pressure Estimation
• EMCee: Improving Multilingual Capability of LLMs via Bridging Knowledge and Reasoning with Extracted Synthetic Multilingual Context
• NFIG: Autoregressive Image Generation with Next-Frequency Prediction
• LinEAS: End-to-end Learning of Activation Steering with a Distributional Loss
• Changing Base Without Losing Pace: A GPU-Efficient Alternative to MatMul in DNNs
• Beyond Final Code: A Process-Oriented Error Analysis of Software Development Agents in Real-World GitHub Scenarios
• Text2Schema: Filling the Gap in Designing Database Table Structures based on Natural Language
• Unfair Learning: GenAI Exceptionalism and Copyright Law
• Multi-identity Human Image Animation with Structural Video Diffusion
• Interpretable Hybrid-Rule Temporal Point Processes
• SHeaP: Self-Supervised Head Geometry Predictor Learned via 2D Gaussians
• MAYA: Addressing Inconsistencies in Generative Password Guessing through a Unified Benchmark
• FLUKE: A Linguistically-Driven and Task-Agnostic Framework for Robustness Evaluation
• PAD: Phase-Amplitude Decoupling Fusion for Multi-Modal Land Cover Classification
• Scaling Multi Agent Reinforcement Learning for Underwater Acoustic Tracking via Autonomous Vehicles
• Uncertainty Quantification for Prior-Data Fitted Networks using Martingale Posteriors
• OpenEstimate: Evaluating LLMs on Reasoning Under Uncertainty with Real-World Data
• Procedural Game Level Design with Deep Reinforcement Learning
• Towards Error Centric Intelligence I, Beyond Observational Learning
• HugAgent: Evaluating LLMs in Simulating Human-Like Individual Reasoning on Open-Ended Tasks
• WELD: A Large-Scale Longitudinal Dataset of Emotional Dynamics for Ubiquitous Affective Computing
• From Checklists to Clusters: A Homeostatic Account of AGI Evaluation
• Multi-dimensional Data Analysis and Applications Basing on LLM Agents and Knowledge Graph Interactions
• Experience-Driven Exploration for Efficient API-Free AI Agents
• AUGUSTUS: An LLM-Driven Multimodal Agent System with Contextualized User Memory
• WebGen-V Bench: Structured Representation for Enhancing Visual Design in LLM-based Web Generation and Evaluation
• VERITAS: Leveraging Vision Priors and Expert Fusion to Improve Multimodal Data
• Towards Flash Thinking via Decoupled Advantage Policy Optimization
• Advancing Routing-Awareness in Analog ICs Floorplanning
• Corrigibility Transformation: Constructing Goals That Accept Updates
• MARS: Reinforcing Multi-Agent Reasoning of LLMs through Self-Play in Strategic Games
• Adaptive Minds: Empowering Agents with LoRA-as-Tools
• Taming the Judge: Deconflicting AI Feedback for Stable Reinforcement Learning
• Hypergraph Contrastive Sensor Fusion for Multimodal Fault Diagnosis in Induction Motors
• JudgeSQL: Reasoning over SQL Candidates with Weighted Consensus Tournament
• Context-aware deep learning using individualized prior information reduces false positives in disease risk prediction and longitudinal health assessment
• Unleashing Scientific Reasoning for Bio-experimental Protocol Generation via Structured Component-based Reward Mechanism
• Build Your Personalized Research Group: A Multiagent Framework for Continual and Interactive Science Automation
• Direct Preference Optimization with Unobserved Preference Heterogeneity: The Necessity of Ternary Preferences
• Invoice Information Extraction: Methods and Performance Evaluation
• AURA: An Agent Autonomy Risk Assessment Framework
• Towards Relaxed Multimodal Inputs for Gait-based Parkinson's Disease Assessment
• Preliminary Quantitative Study on Explainability and Trust in AI Systems
• Self-evolving expertise in complex non-verifiable subject domains: dialogue as implicit meta-RL
• Demo: Guide-RAG: Evidence-Driven Corpus Curation for Retrieval-Augmented Generation in Long COVID