AI Research News Feeds for October 10th, 2025

AI RESEARCH PAPERS & ACADEMIC SOURCES

• NaViL: Rethinking Scaling Properties of Native Multimodal Large Language Models under Data Constraints
• D$^2$GS: Depth-and-Density Guided Gaussian Splatting for Stable and Accurate Sparse-View Reconstruction
• ReSplat: Learning Recurrent Gaussian Splats
• FlowLensing: Simulating Gravitational Lensing with Flow Matching
• SatFusion: A Unified Framework for Enhancing Satellite IoT Images via Multi-Temporal and Multi-Source Data Fusion
• SViM3D: Stable Video Material Diffusion for Single Image 3D Generation
• Spectral Prefiltering of Neural Fields
• Splat the Net: Radiance Fields with Splattable Neural Primitives
• X2Video: Adapting Diffusion Models for Multimodal Controllable Neural Video Rendering
• R2RGEN: Real-to-Real 3D Data Generation for Spatially Generalized Manipulation
• DexNDM: Closing the Reality Gap for Dexterous In-Hand Rotation via Joint-Wise Neural Dynamics Model
• Scalable Offline Metrics for Autonomous Driving
• PRVR: Partially Relevant Video Retrieval
• I&S-ViT: An Inclusive & Stable Method for Pushing the Limit of Post-Training ViTs Quantization
• Redundant Semantic Environment Filling via Misleading-Learning for Fair Deepfake Detection
• Surfel-based Gaussian Inverse Rendering for Fast and Relightable Dynamic Human Reconstruction from Monocular Video
• Motion Capture from Inertial and Vision Sensors
• CurvNet: Latent Contour Representation and Iterative Data Engine for Curvature Angle Estimation
• MonoGSDF: Exploring Monocular Geometric Cues for Gaussian Splatting-Guided Implicit Surface Reconstruction
• EFSA: Episodic Few-Shot Adaptation for Text-to-Image Retrieval
• Scalable Cosmic AI Inference using Cloud Serverless Computing
• Self-Training with Dynamic Weighting for Robust Gradual Domain Adaptation
• H3DE-Net: Efficient and Accurate 3D Landmark Detection in Medical Imaging
• TransMamba: Fast Universal Architecture Adaption from Transformers to Mamba
• DICEPTION: A Generalist Diffusion Model for Visual Perceptual Tasks
• Uncertainty-Aware Diffusion Guided Refinement of 3D Scenes
• Targetless LiDAR-Camera Calibration with Neural Gaussian Splatting
• DvD: Unleashing a Generative Paradigm for Document Dewarping via Coordinates-based Diffusion Model
• ThinkGeo: Evaluating Tool-Augmented Agents for Remote Sensing Tasks
• MotionSight: Boosting Fine-Grained Motion Understanding in Multimodal LLMs
• IMAGHarmony: Controllable Image Editing with Consistent Object Quantity and Layout
• OASIS: Online Sample Selection for Continual Visual Instruction Tuning
• Feedback Guidance of Diffusion Models
• ManipGPT: Is Affordance Segmentation by Large Vision Models Enough for Articulated Object Manipulation?
• Language learning shapes visual category-selectivity in deep neural networks
• MAMBO: High-Resolution Generative Approach for Mammography Images
• Which Heads Matter for Reasoning? RL-Guided KV Cache Compression
• PATCH: Mitigating PII Leakage in Language Models with Privacy-Aware Targeted Circuit PatcHing
• Who Stole Your Data? A Method for Detecting Unauthorized RAG Theft
• From Keywords to Clusters: AI-Driven Analysis of YouTube Comments to Reveal Election Issue Salience in 2024
• Pseudo2Real: Task Arithmetic for Pseudo-Label Correction in Automatic Speech Recognition
• ReasonEmbed: Enhanced Text Embeddings for Reasoning-Intensive Document Retrieval
• The Visual Iconicity Challenge: Evaluating Vision-Language Models on Sign Language Form-Meaning Mapping
• SliceFine: The Universal Winning-Slice Hypothesis for Pretrained Networks
• Evaluating LLMs' Mathematical Reasoning in Financial Document Question Answering
• ThinkNote: Enhancing Knowledge Integration and Utilization of Large Language Models via Constructivist Cognition Modeling
• Expert-Token Resonance MoE: Bidirectional Routing with Efficiency Affinity-Driven Active Selection
• Med-R$^2$: Crafting Trustworthy LLM Physicians via Retrieval and Reasoning of Evidence-Based Medicine
• Mitigating Forgetting in LLM Fine-Tuning via Low-Perplexity Token Learning
• Examining Multilingual Embedding Models Cross-Lingually Through LLM-Generated Adversarial Examples
• Less is More: Compact Clue Selection for Efficient Retrieval-Augmented Generation Reasoning
• Beyond Single Frames: Can LMMs Comprehend Temporal and Contextual Narratives in Image Sequences?
• Argument Summarization and its Evaluation in the Era of Large Language Models
• Sherkala-Chat: Building a State-of-the-Art LLM for Kazakh in a Moderately Resourced Setting
• DiMA: An LLM-Powered Ride-Hailing Assistant at DiDi
• UniEDU: A Unified Language and Vision Assistant for Education Applications
• Can LLMs Grasp Implicit Cultural Values? Benchmarking LLMs' Cultural Intelligence with CQ-Bench
• Say It Another Way: Auditing LLMs with a User-Grounded Automated Paraphrasing Framework
• What Media Frames Reveal About Stance: A Dataset and Study about Memes in Climate Change Discourse
• UNCLE: Benchmarking Uncertainty Expressions in Long-Form Generation
• FlashDLM: Accelerating Diffusion Language Model Inference via Efficient KV Caching and Guided Diffusion
• FlowNIB: An Information Bottleneck Analysis of Bidirectional vs. Unidirectional Language Models
• From Handwriting to Feedback: Evaluating VLMs and LLMs for AI-Powered Assessment in Indonesian Classrooms
• Language Surgery in Multilingual Large Language Models
• How Grounded is Wikipedia? A Study on Structured Evidential Support and Retrieval
• The Behavioural Translation Style Space: Towards simulating the temporal dynamics of affect, behaviour, and cognition in human translation production
• Can Vision Language Models Infer Human Gaze Direction? A Controlled Study
• Play to Generalize: Learning to Reason Through Game Play
• DynamicEval: Rethinking Evaluation for Dynamic Text-to-Video Synthesis
• Provably Accelerated Imaging with Restarted Inertia and Score-based Image Priors
• D2RA: Dual Domain Regeneration Attack
• PickStyle: Video-to-Video Style Transfer with Context-Style Adapters
• Cross-Modal Attention Guided Unlearning in Vision-Language Models
• MaizeStandCounting (MaSC): Automated and Accurate Maize Stand Counting from UAV Imagery Using Image Processing and Deep Learning
• Quick-CapsNet (QCN): A fast alternative to Capsule Networks
• Rectified-CFG++ for Flow Based Models
• PIT-QMM: A Large Multimodal Model For No-Reference Point Cloud Quality Assessment
• Dual-Stream Alignment for Action Segmentation
• Once Is Enough: Lightweight DiT-Based Video Virtual Try-On via One-Time Garment Appearance Injection
• MONKEY: Masking ON KEY-Value Activation Adapter for Personalization
• Automatic Text Box Placement for Supporting Typographic Design
• Hybrid CNN-BYOL Approach for Fault Detection in Induction Motors Using Thermal Images
• Mutual Learning for Hashing: Unlocking Strong Hash Functions from Weak Supervision
• RePainter: Empowering E-commerce Object Removal via Spatial-matting Reinforcement Learning
• SyncHuman: Synchronizing 2D and 3D Generative Models for Single-view Human Reconstruction
• ComGS: Efficient 3D Object-Scene Composition via Surface Octahedral Probes
• DEGS: Deformable Event-based 3D Gaussian Splatting from RGB and Event Stream
• Demystifying Deep Learning-based Brain Tumor Segmentation with 3D UNets and Explainable AI (XAI): A Comparative Analysis
• GTR-Bench: Evaluating Geo-Temporal Reasoning in Vision-Language Models
• FMANet: A Novel Dual-Phase Optical Flow Approach with Fusion Motion Attention Network for Robust Micro-expression Recognition
• An End-to-End Room Geometry Constrained Depth Estimation Framework for Indoor Panorama Images
• Enhancing Visual Prompting through Expanded Transformation Space and Overfitting Mitigation
• MMHOI: Modeling Complex 3D Multi-Human Multi-Object Interactions
• PrismGS: Physically-Grounded Anti-Aliasing for High-Fidelity Large-Scale 3D Gaussian Splatting
• IsoSignVid2Aud: Sign Language Video to Audio Conversion without Text Intermediaries
• AlignGS: Aligning Geometry and Semantics for Robust Indoor Reconstruction from Sparse Views
• XYZCylinder: Feedforward Reconstruction for Driving Scenes Based on A Unified Cylinder Lifting Method
• MARC: Memory-Augmented RL Token Compression for Efficient Video Understanding
• ASBench: Image Anomalies Synthesis Benchmark for Anomaly Detection
• CVD-STORM: Cross-View Video Diffusion with Spatial-Temporal Reconstruction Model for Autonomous Driving
• Latent Harmony: Synergistic Unified UHD Image Restoration via Latent Space Regularization and Controllable Refinement
• The impact of abstract and object tags on image privacy classification
• GraphEnet: Event-driven Human Pose Estimation with a Graph Neural Network
• CIR-CoT: Towards Interpretable Composed Image Retrieval via End-to-End Chain-of-Thought Reasoning
• RayFusion: Ray Fusion Enhanced Collaborative Visual Perception
• RASALoRE: Region Aware Spatial Attention with Location-based Random Embeddings for Weakly Supervised Anomaly Detection in Brain MRI Scans
• RetouchLLM: Training-free White-box Image Retouching
• A class-driven hierarchical ResNet for classification of multispectral remote sensing images
• Towards Real-World Deepfake Detection: A Diverse In-the-wild Dataset of Forgery Faces
• DarkHash: A Data-Free Backdoor Attack Against Deep Hashing
• Efficient Label Refinement for Face Parsing Under Extreme Poses Using 3D Gaussian Splatting
• Real-Time Motion-Controllable Autoregressive Video Diffusion
• UniMMVSR: A Unified Multi-Modal Framework for Cascaded Video Super-Resolution
• Beyond Textual CoT: Interleaved Text-Image Chains with Deep Confidence Reasoning for Image Editing
• InstructUDrag: Joint Text Instructions and Object Dragging for Interactive Image Editing
• Fine-grained text-driven dual-human motion generation via dynamic hierarchical interaction
• Adaptive Gradient Calibration for Single-Positive Multi-Label Learning in Remote Sensing Image Scene Classification
• One Stone with Two Birds: A Null-Text-Null Frequency-Aware Diffusion Models for Text-Guided Image Inpainting
• A Multimodal Depth-Aware Method For Embodied Reference Understanding
• LTCA: Long-range Temporal Context Attention for Referring Video Object Segmentation
• Unlocking 3D Affordance Segmentation with 2D Semantic Knowledge
• LinVideo: A Post-Training Framework towards O(n) Attention in Efficient Video Generation
• SPICE: Simple and Practical Image Clarification and Enhancement
• Hyperspectral data augmentation with transformer-based diffusion models
• UniVideo: Unified Understanding, Generation, and Editing for Videos
• Robust Source-Free Domain Adaptation for Medical Image Segmentation based on Curriculum Learning
• VideoVerse: How Far is Your T2V Generator from a World Model?
• Hierarchical Spatial Algorithms for High-Resolution Image Quantization and Feature Extraction
• Video-STAR: Reinforcing Open-Vocabulary Action Recognition with Tools
• InstructX: Towards Unified Visual Editing with MLLM Guidance
• MoA-VR: A Mixture-of-Agents System Towards All-in-One Video Restoration
• Have We Scene It All? Scene Graph-Aware Deep Point Cloud Compression
• FlexTraj: Image-to-Video Generation with Flexible Point Trajectory Control
• MM-HELIX: Boosting Multimodal Long-Chain Reflective Reasoning with Holistic Platform and Adaptive Hybrid Policy Optimization
• ARTDECO: Towards Efficient and High-Fidelity On-the-Fly 3D Reconstruction with Structured Scene Representation
• VideoCanvas: Unified Video Completion from Arbitrary Spatiotemporal Patches via In-Context Conditioning
• MultiCOIN: Multi-Modal COntrollable Video INbetweening
• ResAD: Normalized Residual Trajectory Modeling for End-to-End Autonomous Driving
• Lemma Dilemma: On Lemma Generation Without Domain- or Language-Specific Training Data
• Meaningful Pose-Based Sign Language Evaluation
• Populism Meets AI: Advancing Populism Research with LLMs
• MAPRO: Recasting Multi-Agent Prompt Optimization as Maximum a Posteriori Inference
• AsyncSpade: Efficient Test-Time Scaling with Asynchronous Sparse Decoding
• ParsTranslit: Truly Versatile Tajik-Farsi Transliteration
• IASC: Interactive Agentic System for ConLangs
• Toward Reliable Clinical Coding with Language Models: Verification and Lightweight Adaptation
• Role-Conditioned Refusals: Evaluating Access Control Reasoning in Large Language Models
• Textual Entailment and Token Probability as Bias Evaluation Metrics
• MemWeaver: A Hierarchical Memory from Textual Interactive Behaviors for Personalized Generation
• SUBQRAG: sub-question driven dynamic graph rag
• Multilingual Knowledge Graph Completion via Efficient Multilingual Knowledge Sharing
• OpenRubrics: Towards Scalable Synthetic Rubric Generation for Reward Modeling and LLM Alignment
• Test-Time Reasoners Are Strategic Multiple-Choice Test-Takers
• Curing Miracle Steps in LLM Mathematical Reasoning with Rubric Rewards
• The Unintended Trade-off of AI Alignment:Balancing Hallucination Mitigation and Safety in LLMs
• RCPU: Rotation-Constrained Error Compensation for Structured Pruning of a Large Language Model
• Multilingual Generative Retrieval via Cross-lingual Semantic Compression
• Ready to Translate, Not to Represent? Bias and Performance Gaps in Multilingual LLMs Across Language Families and Domains
• Do LLMs Really Need 10+ Thoughts for "Find the Time 1000 Days Later"? Towards Structural Understanding of LLM Overthinking
• CS3-Bench: Evaluating and Enhancing Speech-to-Speech LLMs for Mandarin-English Code-Switching
• Standard-to-Dialect Transfer Trends Differ across Text and Speech: A Case Study on Intent and Topic Classification in German Dialects
• Metric Calculating Benchmark: Code-Verifiable Complicate Instruction Following Benchmark for Large Language Models
• ACE: Attribution-Controlled Knowledge Editing for Multi-hop Factual Recall
• Comprehensiveness Metrics for Automatic Evaluation of Factual Recall in Text Generation
• Vision-Enabled LLMs in Historical Lexicography: Digitising and Enriching Estonian-German Dictionaries from the 17th and 18th Centuries
• ChatGPT as a Translation Engine: A Case Study on Japanese-English
• Evaluating LLM-Generated Legal Explanations for Regulatory Compliance in Social Media Influencer Marketing
• Mitigating Judgment Preference Bias in Large Language Models through Group-Based Polling
• Beyond Over-Refusal: Scenario-Based Diagnostics and Post-Hoc Mitigation for Exaggerated Refusals in LLMs
• ARM2: Adaptive Reasoning Model with Vision Understanding and Executable Code
• METRICALARGS: A Taxonomy for Studying Metrical Poetry with LLMs
• Training-Free Group Relative Policy Optimization
• SenWave: A Fine-Grained Multi-Language Sentiment Analysis Dataset Sourced from COVID-19 Tweets
• The Alignment Waltz: Jointly Training Agents to Collaborate for Safety
• Beyond Turn Limits: Training Deep Search Agents with Dynamic Context Window
• Neuron-Level Analysis of Cultural Understanding in Large Language Models
• AutoRed: A Free-form Adversarial Prompt Generation Framework for Automated Red Teaming
• Two-Stage Voting for Robust and Efficient Suicide Risk Detection on Social Media
• If Probable, Then Acceptable? Understanding Conditional Acceptability Judgments in Large Language Models
• ARES: Multimodal Adaptive Reasoning via Difficulty-Aware Token-Level Entropy Shaping
• LeWiDi-2025 at NLPerspectives: The Third Edition of the Learning with Disagreements Shared Task
• Neologism Learning for Controllability and Self-Verbalization
• Efficient Prompt Optimisation for Legal Text Classification with Proxy Prompt Evaluator
• WebAgent-R1: Training Web Agents via End-to-End Multi-Turn Reinforcement Learning
• LLM Fingerprinting via Semantically Conditioned Watermarks
• Continuum Transformers Perform In-Context Learning by Operator Gradient Descent
• Scaling Laws Are Unreliable for Downstream Tasks: A Reality Check
• Multi-Trigger Poisoning Amplifies Backdoor Vulnerabilities in LLMs
• Zebra-CoT: A Dataset for Interleaved Vision Language Reasoning
• Arbitrary Entropy Policy Optimization: Entropy Is Controllable in Reinforcement Finetuning
• Unsupervised Multi-Source Federated Domain Adaptation under Domain Diversity through Group-Wise Discrepancy Minimization
• Beyond Sub-6 GHz: Leveraging mmWave Wi-Fi for Gait-Based Person Identification
• Bidirectional Representations Augmented Autoregressive Biological Sequence Generation:Application in De Novo Peptide Sequencing
• Long-tailed Recognition with Model Rebalancing
• Dual-granularity Sinkhorn Distillation for Enhanced Learning from Long-tailed Noisy Data
• Post-hoc Stochastic Concept Bottleneck Models
• Reinforcement Learning from Probabilistic Forecasts for Safe Decision-Making via Conditional Value-at-Risk Planning
• Enhancing Reasoning for Diffusion LLMs via Distribution Matching Policy Optimization
• Bridging the Physics-Data Gap with FNO-Guided Conditional Flow Matching: Designing Inductive Bias through Hierarchical Physical Constraints
• Dynamic Features Adaptation in Networking: Toward Flexible training and Explainable inference
• Robust and Efficient Collaborative Learning
• To Ask or Not to Ask: Learning to Require Human Feedback
• Guided Star-Shaped Masked Diffusion
• Contrastive Self-Supervised Learning at the Edge: An Energy Perspective
• Characterizing the Multiclass Learnability of Forgiving 0-1 Loss Functions
• Biology-driven assessment of deep learning super-resolution imaging of the porosity network in dentin
• Reinforcing Diffusion Models by Direct Group Preference Optimization
• SummDiff: Generative Modeling of Video Summarization with Diffusion
• In-Context Clustering with Large Language Models
• Better Together: Leveraging Unpaired Multimodal Data for Stronger Unimodal Models
• DYNAMIX: RL-based Adaptive Batch Size Optimization in Distributed Machine Learning Systems
• Convergence Theorems for Entropy-Regularized and Distributional Reinforcement Learning
• Entropy Regularizing Activation: Boosting Continuous Control, Large Language Models, and Image Classification with Activation as Entropy Constraints
• Improving Reasoning for Diffusion Language Models via Group Diffusion Policy Optimization
• Who Said Neural Networks Aren't Linear?
• Geodesics in the Deep Linear Network
• Decoding the dark proteome: Deep learning-enabled discovery of druggable enzymes in Wuchereria bancrofti
• SpotDiff: Spotting and Disentangling Interference in Feature Space for Subject-Preserving Image Generation
• Beyond Grid-Locked Voxels: Neural Response Functions for Continuous Brain Encoding
• Enhancing Maritime Object Detection in Real-Time with RT-DETR and Data Augmentation
• Inconsistent Affective Reaction: Sentiment of Perception and Opinion in Urban Environments
• Bayesian Optimization of Multi-Bit Pulse Encoding in In2O3/Al2O3 Thin-film Transistors for Temporal Data Processing
• VeMo: A Lightweight Data-Driven Approach to Model Vehicle Dynamics
• Comparison of Fully Homomorphic Encryption and Garbled Circuit Techniques in Privacy-Preserving Machine Learning Inference
• Evaluating and Learning Optimal Dynamic Treatment Regimes under Truncation by Death
• Time-Frequency Filtering Meets Graph Clustering
• Beyond independent component analysis: identifiability and algorithms
• Deploying Tiny LVLM Judges for Real-World Evaluation of Chart Models: Lessons Learned and Best Practices
• Locality-Sensitive Hashing-Based Efficient Point Transformer for Charged Particle Reconstruction
• From Data to Rewards: a Bilevel Optimization Perspective on Maximum Likelihood Estimation
• A Honest Cross-Validation Estimator for Prediction Performance
• Large Language Models Meet Virtual Cell: A Survey
• ToolExpander: Extending the Frontiers of Tool-Using Reinforcement Learning to Weak LLMs
• When Robustness Meets Conservativeness: Conformalized Uncertainty Calibration for Balanced Decision Making
• Instance Relation Learning Network with Label Knowledge Propagation for Few-shot Multi-label Intent Detection
• PLUM: Adapting Pre-trained Language Models for Industrial-scale Generative Recommendations
• Adaptive Execution Scheduler for DataDios SmartDiff
• Surrogate Graph Partitioning for Spatial Prediction
• On the Optimality of Tracking Fisher Information in Adaptive Testing with Stochastic Binary Responses
• On the Optimality of the Median-of-Means Estimator under Adversarial Contamination
• Multi-level informed optimization via decomposed Kriging for large design problems under uncertainty
• SimCast: Enhancing Precipitation Nowcasting with Short-to-Long Term Knowledge Distillation
• Stick-Breaking Mixture Normalizing Flows with Component-Wise Tail Adaptation for Variational Inference
• Climate Knowledge in Large Language Models
• Physics-Driven Spatiotemporal Modeling for AI-Generated Video Detection
• Detecting and Mitigating Insertion Hallucination in Video-to-Audio Generation
• Computations and ML for surjective rational maps
• Beyond Real Data: Synthetic Data through the Lens of Regularization
• Random Window Augmentations for Deep Learning Robustness in CT and Liver Tumor Segmentation
• High-dimensional Analysis of Synthetic Data Selection
• Investigating Counterclaims in Causality Extraction from Text
• New Machine Learning Approaches for Intrusion Detection in ADS-B
• PAC Learnability in the Presence of Performativity
• On the Relationship Between the Choice of Representation and In-Context Learning
• Optimal Stopping in Latent Diffusion Models
• Large Scale Diffusion Distillation via Score-Regularized Continuous-Time Consistency
• Navigating Sparsities in High-Dimensional Linear Contextual Bandits
• Wavefunction Flows: Efficient Quantum Simulation of Continuous Flow Models
• Don't Run with Scissors: Pruning Breaks VLA Models but They Can Be Recovered
• Accelerated Aggregated D-Optimal Designs for Estimating Main Effects in Black-Box Models
• DexMan: Learning Bimanual Dexterous Manipulation from Human and Generated Videos
• Implementing Semantic Join Operators Efficiently
• Permutation-Invariant Spectral Learning via Dyson Diffusion
• Computational and statistical lower bounds for low-rank estimation under general inhomogeneous noise
• SPAD: Specialized Prefill and Decode Hardware for Disaggregated LLM Inference
• Where Have All the Kaczmarz Iterates Gone?
• Reconstructing the local density field with combined convolutional and point cloud architecture
• Maintaining Performance with Less Data
• Stochastic Interpolants: A Unifying Framework for Flows and Diffusions
• Graph-SCP: Accelerating Set Cover Problems with Graph Neural Networks
• The Poisson Midpoint Method for Langevin Dynamics: Provably Efficient Discretization for Diffusion Models
• Adaptive Collaborative Correlation Learning-based Semi-Supervised Multi-Label Feature Selection
• Mitigating Noise Detriment in Differentially Private Federated Learning with Model Pre-training
• PFAttack: Stealthy Attack Bypassing Group Fairness in Federated Learning
• Personalized Federated Fine-Tuning for LLMs via Data-Driven Heterogeneous Model Architectures
• Empirical evaluation of normalizing flows in Markov Chain Monte Carlo
• Efficient Graph Condensation via Gaussian Process
• Learning General Causal Structures with Hidden Dynamic Process for Climate Analysis
• Task Vector Bases: A Unified and Scalable Framework for Compressed Task Arithmetic
• InfoPos: A Design Support Framework for ML-Assisted Fault Detection and Identification in Industrial Cyber-Physical Systems
• Uncertainty Comes for Free: Human-in-the-Loop Policies with Diffusion Models
• Learn to Bid as a Price-Maker Wind Power Producer
• Unified Cross-Scale 3D Generation and Understanding via Autoregressive Modeling
• Solving Time-Fractional Partial Integro-Differential Equations Using Tensor Neural Network
• Chisme: Fully Decentralized Differentiated Deep Learning for IoT Intelligence
• Learning Equilibria from Data: Provably Efficient Multi-Agent Imitation Learning
• Can Large Reasoning Models Self-Train?
• Martingale Posterior Neural Networks for Fast Sequential Decision Making
• Little By Little: Continual Learning via Self-Activated Sparse Mixture-of-Rank Adaptive Learning
• Anticipating the Selectivity of Intramolecular Cyclization Reaction Pathways with Neural Network Potentials
• Cost-aware Stopping for Bayesian Optimization
• A Kernel Distribution Closeness Testing
• HyPINO: Multi-Physics Neural Operators via HyperPINNs and the Method of Manufactured Solutions
• TiAda: A Time-scale Adaptive Algorithm for Nonconvex Minimax Optimization
• It's All in the Mix: Wasserstein Classification and Regression with Mixed Features
• Attention based End to end network for Offline Writer Identification on Word level data
• Data-Error Scaling Laws in Machine Learning on Combinatorial Mutation-prone Sets: Proteins and Small Molecules
• Recurrent Natural Policy Gradient for POMDPs
• MeanSparse: Post-Training Robustness Enhancement Through Mean-Centered Feature Sparsification
• BaTCAVe: Trustworthy Explanations for Robot Behaviors
• Latency-Aware Contextual Bandit: Application to Cryo-EM Data Collection
• Rethinking Decoders for Transformer-based Semantic Segmentation: A Compression Perspective
• Learning to Partially Defer for Sequences
• Distribution Transformers: Fast Approximate Bayesian Inference With On-The-Fly Prior Adaptation
• Erasing Without Remembering: Implicit Knowledge Forgetting in Large Language Models
• Markets for Models
• Efficient and Adaptable Overlapping for Computation and Communication via Signaling and Reordering
• Phantora: Maximizing Code Reuse in Simulation-based Machine Learning System Performance Estimation
• Efficient Multi Subject Visual Reconstruction from fMRI Using Aligned Representations
• SmartUT: Receive Beamforming for Spectral Coexistence of NGSO Satellite Systems
• Graphon Mixtures
• PO-Flow: Flow-based Generative Models for Sampling Potential Outcomes and Counterfactuals
• Foundation Models for Structural Health Monitoring
• Objective Features Extracted from Motor Activity Time Series for Food Addiction Analysis Using Machine Learning - A Pilot Study
• Nearest Neighbor CCP-Based Molecular Sequence Analysis
• Multi-Source Knowledge Pruning for Retrieval-Augmented Generation: A Benchmark and Empirical Study
• Language Model Embeddings Can Be Sufficient for Bayesian Optimization
• Multi-Continental Healthcare Modelling Using Blockchain-Enabled Federated Learning
• Matryoshka Pilot: Learning to Drive Black-Box LLMs with LLMs
• RAGDiffusion: Faithful Cloth Generation via External Knowledge Assimilation
• Kernel-Free Universum Quadratic Surface Twin Support Vector Machines for Imbalanced Data
• HiVeGen -- Hierarchical LLM-based Verilog Generation for Scalable Chip Design
• EpiCoder: Encompassing Diversity and Complexity in Code Generation
• BRIGHT: A globally distributed multimodal building damage assessment dataset with very-high-resolution for all-weather disaster response
• Self-Improving Skill Learning for Robust Skill-based Meta-Reinforcement Learning
• Rex: Reversible Solvers for Diffusion Models
• MoM: Linear Sequence Modeling with Mixture-of-Memories
• BixBench: a Comprehensive Benchmark for LLM-based Agents in Computational Biology
• Enhancing LLM Reliability via Explicit Knowledge Boundary Modeling
• LLM Applications: Current Paradigms and the Next Frontier
• Adoption of Watermarking for Generative AI Systems in Practice and Implications under the new EU AI Act
• More Bang for the Buck: Process Reward Modeling with Entropy-Driven Uncertainty
• Adaptive Layer-skipping in Pre-trained LLMs
• $\textit{Agents Under Siege}$: Breaking Pragmatic Multi-Agent LLM Systems with Optimized Prompt Attacks
• PiCo: Jailbreaking Multimodal Large Language Models via Pictorial Code Contextualization
• Hallucination Detection in LLMs with Topological Divergence on Attention Graphs
• T-VEC: A Telecom-Specific Vectorization Model with Enhanced Semantic Understanding via Deep Triplet Loss Fine-Tuning
• Evaluating Evaluation Metrics -- The Mirage of Hallucination Detection
• Understanding In-context Learning of Addition via Activation Subspaces
• Hakim: Farsi Text Embedding Model
• FairSHAP: Preprocessing for Fairness Through Attribution-Based Data Augmentation
• Logic Jailbreak: Efficiently Unlocking LLM Safety Restrictions Through Formal Logical Expression
• LLINBO: Trustworthy LLM-in-the-Loop Bayesian Optimization
• Watch your steps: Dormant Adversarial Behaviors that Activate upon LLM Finetuning
• Search Wisely: Mitigating Sub-optimal Agentic Searches By Reducing Uncertainty
• STOPA: A Database of Systematic VariaTion Of DeePfake Audio for Open-Set Source Tracing and Attribution
• Inference-time Alignment in Continuous Space
• The Shape of Adversarial Influence: Characterizing LLM Latent Spaces with Persistent Homology
• Trans-EnV: A Framework for Evaluating the Linguistic Robustness of LLMs Against English Varieties
• CAST: Contrastive Adaptation and Distillation for Semi-Supervised Instance Segmentation
• GL-PGENet: A Parameterized Generation Framework for Robust Document Image Enhancement
• MAGREF: Masked Guidance for Any-Reference Video Generation with Subject Disentanglement
• Tug-of-war between idioms' figurative and literal interpretations in LLMs
• Dissecting Logical Reasoning in LLMs: A Fine-Grained Evaluation and Supervision Study
• Modality-Balancing Preference Optimization of Large Multimodal Models by Adversarial Negative Mining
• Product of Experts for Visual Generation
• Intention-Conditioned Flow Occupancy Models
• ReasonMed: A 370K Multi-Agent Generated Dataset for Advancing Medical Reasoning
• Rethinking Losses for Diffusion Bridge Samplers
• Not All Clients Are Equal: Collaborative Model Personalization on Heterogeneous Multi-Modal Clients
• Breaking the Reviewer: Assessing the Vulnerability of Large Language Models in Automated Peer Review Under Textual Adversarial Attacks
• The Role of Model Confidence on Bias Effects in Measured Uncertainties for Vision-Language Models
• LLMs on a Budget? Say HOLA
• Truth, Trust, and Trouble: Medical AI on the Edge
• Efficiency-Effectiveness Reranking FLOPs for LLM-based Rerankers
• ERR@HRI 2.0 Challenge: Multimodal Detection of Errors and Failures in Human-Robot Conversations
• Understanding Teen Overreliance on AI Companion Chatbots Through Self-Reported Reddit Narratives
• Leveraging Personalized PageRank and Higher-Order Topological Structures for Heterophily Mitigation in Graph Neural Networks
• A Modality-Aware Cooperative Co-Evolutionary Framework for Multimodal Graph Neural Architecture Search
• Out-of-Distribution Generalization in Climate-Aware Yield Prediction with Earth Observation Data
• ConCuR: Conciseness Makes State-of-the-Art Kernel Generation
• Best-of-Both Worlds for linear contextual bandits with paid observations
• Learning to Route LLMs from Bandit Feedback: One Policy, Many Trade-offs
• Parameter-Free Federated TD Learning with Markov Noise in Heterogeneous Environments
• metabeta - A fast neural model for Bayesian mixed-effects regression
• Surrogate Modeling for the Design of Optimal Lattice Structures using Tensor Completion
• Reinforcement Learning-based Task Offloading in the Internet of Wearable Things
• Black-box Detection of LLM-generated Text Using Generalized Jensen-Shannon Divergence
• PEAR: Planner-Executor Agent Robustness Benchmark
• Efficient Generalization via Multimodal Co-Training under Data Scarcity and Distribution Shift
• Estimating Fair Graphs from Graph-Stationary Data
• Targeted Digital Twin via Flow Map Learning and Its Application to Fluid Dynamics
• Phase Diagram of Dropout for Two-Layer Neural Networks in the Mean-Field Regime
• EBGAN-MDN: An Energy-Based Adversarial Framework for Multi-Modal Behavior Cloning
• Automated Machine Learning for Unsupervised Tabular Tasks
• Symbolic-Diffusion: Deep Learning Based Symbolic Regression with D3PM Discrete Token Diffusion
• Expanding the Action Space of LLMs to Reason Beyond Language
• Transformer-Based Indirect Structural Health Monitoring of Rail Infrastructure with Attention-Driven Detection and Localization of Transient Defects
• LLM Unlearning Under the Microscope: A Full-Stack View on Methods and Metrics
• Property Classification of Vacation Rental Properties during Covid-19
• Design-Based Bandits Under Network Interference: Trade-Off Between Regret and Statistical Inference
• Continual Learning for Adaptive AI Systems
• Incremental Hybrid Ensemble with Graph Attention and Frequency-Domain Features for Stable Long-Term Credit Risk Modeling
• FedQS: Optimizing Gradient and Model Aggregation for Semi-Asynchronous Federated Learning
• LiveThinking: Enabling Real-Time Efficient Reasoning for AI-Powered Livestreaming via Reinforcement Learning
• Computationally-efficient Graph Modeling with Refined Graph Random Features
• GeoGen: A Two-stage Coarse-to-Fine Framework for Fine-grained Synthetic Location-based Social Network Trajectory Generation
• t-SNE Exaggerates Clusters, Provably
• FedBook: A Unified Federated Graph Foundation Codebook with Intra-domain and Inter-domain Knowledge Modeling
• R\'enyi Sharpness: A Novel Sharpness that Strongly Correlates with Generalization
• FedLAM: Low-latency Wireless Federated Learning via Layer-wise Adaptive Modulation
• Weak Form Learning for Mean-Field Partial Differential Equations: an Application to Insect Movement
• HySim-LLM: Embedding-Weighted Fine-Tuning Bounds and Manifold Denoising for Domain-Adapted LLMs
• Signal-to-Noise Ratio in Scanning Electron Microscopy: A Comprehensive Review
• Adaptive Optimizable Gaussian Process Regression Linear Least Squares Regression Filtering Method for SEM Images
• GRADE: Personalized Multi-Task Fusion via Group-relative Reinforcement Learning with Adaptive Dirichlet Exploratio
• SketchGuard: Scaling Byzantine-Robust Decentralized Federated Learning via Sketch-Based Screening
• Synergy Between the Strong and the Weak: Spiking Neural Networks are Inherently Self-Distillers
• Some theoretical improvements on the tightness of PAC-Bayes risk certificates for neural networks
• PRESCRIBE: Predicting Single-Cell Responses with Bayesian Estimation
• Climate Surrogates for Scalable Multi-Agent Reinforcement Learning: A Case Study with CICERO-SCM
• DemandCast: Global hourly electricity demand forecasting
• Recycling Pretrained Checkpoints: Orthogonal Growth of Mixture-of-Experts for Efficient Large Language Model Pre-Training
• Accelerated Evolving Set Processes for Local PageRank Computation
• Unsupervised Radio Map Construction in Mixed LoS/NLoS Indoor Environments
• Do We Really Need Permutations? Impact of Width Expansion on Linear Mode Connectivity
• From Tokens to Layers: Redefining Stall-Free Scheduling for LLM Serving with Layered Prefill
• Mitigating Subject Dependency in EEG Decoding with Subject-Specific Low-Rank Adapters
• Trajectory Conditioned Cross-embodiment Skill Transfer
• Drift No More? Context Equilibria in Multi-Turn LLM Interactions
• IntentionVLA: Generalizable and Efficient Embodied Intention Reasoning for Human-Robot Interaction
• LLM4Cell: A Survey of Large Language and Agentic Models for Single-Cell Biology
• HiPRAG: Hierarchical Process Rewards for Efficient Agentic Retrieval Augmented Generation
• Dynamic Generation of Multi-LLM Agents Communication Topologies with Graph Diffusion Models
• Effective and Stealthy One-Shot Jailbreaks on Deployed Mobile Vision-Language Agents
• SIMU: Selective Influence Machine Unlearning
• The Rise of the Knowledge Sculptor: A New Archetype for Knowledge Work in the Age of Generative AI
• MetaDefense: Defending Finetuning-based Jailbreak Attack Before and During Generation
• Self-Improving LLM Agents at Test-Time
• AdaSwitch: Adaptive Switching Generation for Knowledge Distillation
• Meta-Learning Based Few-Shot Graph-Level Anomaly Detection
• Self-Supervised Learning Strategies for a Platform to Test the Toxicity of New Chemicals and Materials
• DM1: MeanFlow with Dispersive Regularization for 1-Step Robotic Manipulation
• Team Xiaomi EV-AD VLA: Learning to Navigate Socially Through Proactive Risk Perception - Technical Report for IROS 2025 RoboSense Challenge Social Navigation Track
• Contrastive Weak-to-strong Generalization
• MMM: Quantum-Chemical Molecular Representation Learning for Combinatorial Drug Recommendation
• Towards Human-Like Grading: A Unified LLM-Enhanced Framework for Subjective Question Evaluation
• STEPER: Step-wise Knowledge Distillation for Enhancing Reasoning Ability in Multi-Step Retrieval-Augmented Language Models
• TTOM: Test-Time Optimization and Memorization for Compositional Video Generation
• A Large-scale Dataset for Robust Complex Anime Scene Text Detection
• A$^2$Search: Ambiguity-Aware Question Answering with Reinforcement Learning
• DISCO: Diversifying Sample Condensation for Efficient Model Evaluation
• A Systematic Evaluation of Self-Supervised Learning for Label-Efficient Sleep Staging with Wearable EEG
• LightReasoner: Can Small Language Models Teach Large Language Models Reasoning?
• Active Confusion Expression in Large Language Models: Leveraging World Models toward Better Social Reasoning
• Executable Analytic Concepts as the Missing Link Between VLM Insight and Precise Manipulation
• Unveiling the Power of Multiple Gossip Steps: A Stability-Based Generalization Analysis in Decentralized Training
• ZeroCard: Cardinality Estimation with Zero Dependence on Target Databases -- No Data, No Query, No Retraining
• Is Architectural Complexity Always the Answer? A Case Study on SwinIR vs. an Efficient CNN
• Fewer Weights, More Problems: A Practical Attack on LLM Pruning
• Leveraging Author-Specific Context for Scientific Figure Caption Generation: 3rd SciCap Challenge
• Learning on the Job: An Experience-Driven Self-Evolving Agent for Long-Horizon Tasks
• Past, Present, and Future of Bug Tracking in the Generative AI Era
• Backdoor Vectors: a Task Arithmetic View on Backdoor Attacks and Defenses
• FastUMI-100K: Advancing Data-driven Robotic Manipulation with a Large-scale UMI-style Dataset
• MRI-derived quantification of hepatic vessel-to-volume ratios in chronic liver disease using a deep learning approach
• Towards Reliable LLM-based Robot Planning via Combined Uncertainty Estimation
• Verifying Graph Neural Networks with Readout is Intractable
• TaoSR-AGRL: Adaptive Guided Reinforcement Learning Framework for E-commerce Search Relevance
• A Survey of Process Reward Models: From Outcome Signals to Process Supervisions for Large Language Models
• FedDTRE: Federated Dialogue Generation Models Powered by Trustworthiness Evaluation
• Attribution-by-design: Ensuring Inference-Time Provenance in Generative Music Systems
• An Adaptive Multi Agent Bitcoin Trading System
• A Novel Ensemble Learning Approach for Enhanced IoT Attack Detection: Redefining Security Paradigms in Connected Systems
• Everything is Plausible: Investigating the Impact of LLM Rationales on Human Notions of Plausibility
• The Price of Thought: A Multilingual Analysis of Reasoning, Performance, and Cost of Negotiation in Large Language Models
• Lossless Vocabulary Reduction for Auto-Regressive Language Models
• Development of Mental Models in Human-AI Collaboration: A Conceptual Framework
• VersionRAG: Version-Aware Retrieval-Augmented Generation for Evolving Documents
• Bayesian Decision Making around Experts
• Interpreting LLM-as-a-Judge Policies via Verifiable Global Explanations
• Approximate Domain Unlearning for Vision-Language Models
• Improving Temporal Understanding Logic Consistency in Video-Language Models via Attention Enhancement
• Think Just Enough: Sequence-Level Entropy as a Confidence Signal for LLM Reasoning
• AI Knowledge Assist: An Automated Approach for the Creation of Knowledge Bases for Conversational AI Agents
• DACIP-RC: Domain Adaptive Continual Instruction Pre-Training via Reading Comprehension on Business Conversations
• Quantum Agents for Algorithmic Discovery
• NavSpace: How Navigation Agents Follow Spatial Intelligence Instructions
• Leveraging Whisper Embeddings for Audio-based Lyrics Matching
• Robust Canonicalization through Bootstrapped Data Re-Alignment
• Sentiment Matters: An Analysis of 200 Human-SAV Interactions
• Memory Retrieval and Consolidation in Large Language Models through Function Tokens
• LLMs Learn to Deceive Unintentionally: Emergent Misalignment in Dishonesty from Misaligned Samples to Biased Human-AI Interactions
• FuelCast: Benchmarking Tabular and Temporal Models for Ship Fuel Consumption
• Expressive Value Learning for Scalable Offline Reinforcement Learning
• The Hidden Bias: A Study on Explicit and Implicit Political Stereotypes in Large Language Models
• Contrastive Decoding for Synthetic Data Generation in Low-Resource Language Modeling
• Opponent Shaping in LLM Agents
• Mix- and MoE-DPO: A Variational Inference Approach to Direct Preference Optimization
• A Distributed Emulation Environment for In-Memory Computing Systems
• Learning Neural Exposure Fields for View Synthesis
• Counterfactual Identifiability via Dynamic Optimal Transport
• Iterated Agent for Symbolic Regression
• Learning What's Missing: Attention Dispersion and EMA Stabilization in Length Generalization
• DeepEN: Personalized Enteral Nutrition for Critically Ill Patients using Deep Reinforcement Learning
• Evaluating Small Vision-Language Models on Distance-Dependent Traffic Perception
• Airy: Reading Robot Intent through Height and Sky
• Detecting Legend Items on Historical Maps Using GPT-4o with In-Context Learning
• FlyLoRA: Boosting Task Decoupling and Parameter Efficiency via Implicit Rank-Wise Mixture-of-Experts
• Single layer tiny Co$^4$ outpaces GPT-2 and GPT-BERT
• Prompts Generalize with Low Data: Non-vacuous Generalization Bounds for Optimizing Prompts with More Informative Priors
• ClauseLens: Clause-Grounded, CVaR-Constrained Reinforcement Learning for Trustworthy Reinsurance Pricing
• xRouter: Training Cost-Aware LLMs Orchestration System via Reinforcement Learning
• Gaze on the Prize: Shaping Visual Attention with Return-Guided Contrastive Learning
• Synthetic Series-Symbol Data Generation for Time Series Foundation Models
• gLSTM: Mitigating Over-Squashing by Increasing Storage Capacity
• Integral Signatures of Activation Functions: A 9-Dimensional Taxonomy and Stability Theory for Deep Learning
• Platform-Agnostic Modular Architecture for Quantum Benchmarking
• DeepPrune: Parallel Scaling without Inter-trace Redundancy
• AI-Driven Radiology Report Generation for Traumatic Brain Injuries
• To Sink or Not to Sink: Visual Information Pathways in Large Vision-Language Models
• CoMAS: Co-Evolving Multi-Agent Systems via Interaction Rewards
• SpatialLadder: Progressive Training for Spatial Reasoning in Vision-Language Models
• Kontinuous Kontext: Continuous Strength Control for Instruction-based Image Editing
• On the optimization dynamics of RLVR: Gradient gap and step size thresholds
• VideoNorms: Benchmarking Cultural Awareness of Video Language Models
• Dream to Recall: Imagination-Guided Experience Retrieval for Memory-Persistent Vision-and-Language Navigation
• SciVideoBench: Benchmarking Scientific Video Reasoning in Large Multimodal Models
• MATRIX: Multimodal Agent Tuning for Robust Tool-Use Reasoning
• NovaFlow: Zero-Shot Manipulation via Actionable Flow from Generated Videos
• ArenaBencher: Automatic Benchmark Evolution via Multi-Model Competitive Evaluation
• BLAZER: Bootstrapping LLM-based Manipulation Agents with Zero-Shot Data Generation
• Advancing Automated Urban Planning: Exploring Algorithmic Approaches with Generative Artificial Intelligence
• LogicMP: A Neuro-symbolic Approach for Encoding First-order Logic Constraints
• Average Controlled and Average Natural Micro Direct Effects in Summary Causal Graphs
• Aligning LLM+PDDL Symbolic Plans with Human Objective Specifications through Evolutionary Algorithm Guidance
• BFS-Prover: Scalable Best-First Tree Search for LLM-based Automatic Theorem Proving
• AutoAgent: A Fully-Automated and Zero-Code Framework for LLM Agents
• Position Paper: Towards Open Complex Human-AI Agents Collaboration Systems for Problem Solving and Knowledge Management
• Advancing AI Research Assistants with Expert-Involved Learning
• Adaptive Frontier Exploration on Graphs with Applications to Network-Based Disease Testing
• Let's Reason Formally: Natural-Formal Hybrid Reasoning Enhances LLM's Math Capability
• Bloated Disclosures: Can ChatGPT Help Investors Process Information?
• Contrastive Difference Predictive Coding
• Ultra-Efficient On-Device Object Detection on AI-Integrated Smart Glasses with TinyissimoYOLO
• Thousands of AI Authors on the Future of AI
• Depression Detection on Social Media with Large Language Models
• Truth-Aware Decoding: A Program-Logic Approach to Factual Language Generation
• L2M-AID: Autonomous Cyber-Physical Defense by Fusing Semantic Reasoning of Large Language Models with Multi-Agent Reinforcement Learning (Preprint)
• Base Models Know How to Reason, Thinking Models Learn When
• Position: AI Will Transform Neuropsychology Through Mental Health Digital Twins for Dynamic Mental Health Care, Especially for ADHD
• ProSEA: Problem Solving via Exploration Agents
• Less is More: Strategic Expert Selection Outperforms Ensemble Complexity in Traffic Forecasting
• TS-Agent: A Time Series Reasoning Agent with Iterative Statistical Insight Gathering
• ExpertAgent: Enhancing Personalized Education through Dynamic Planning and Retrieval-Augmented Long-Chain Reasoning
• Evaluation of LLMs for Process Model Analysis and Optimization
• Optimizing Ethical Risk Reduction for Medical Intelligent Systems with Constraint Programming
• CompassLLM: A Multi-Agent Approach toward Geo-Spatial Reasoning for Popular Path Query
• Measuring and Mitigating Identity Bias in Multi-Agent Debate via Anonymization
• An Evaluation Study of Hybrid Methods for Multilingual PII Detection
• Benchmarking is Broken - Don't Let AI be its Own Judge
• AgentAsk: Multi-Agent Systems Need to Ask
• Traceability and Accountability in Role-Specialized Multi-Agent LLM Pipelines
• A Case for Leveraging Generative AI to Expand and Enhance Training in the Provision of Mental Health Services
• Test-Time Matching: Unlocking Compositional Reasoning in Multimodal Models
• Safely Exploring Novel Actions in Recommender Systems via Deployment-Efficient Policy Learning
• Multimodal Safety Evaluation in Generative Agent Social Simulations
• Control Synthesis of Cyber-Physical Systems for Real-Time Specifications through Causation-Guided Reinforcement Learning
• oMeBench: Towards Robust Benchmarking of LLMs in Organic Mechanism Elucidation and Reasoning
• SurveyG: A Multi-Agent LLM Framework with Hierarchical Citation Graph for Automated Survey Generation
• Haibu Mathematical-Medical Intelligent Agent:Enhancing Large Language Model Reliability in Medical Tasks via Verifiable Reasoning Chains
• From Noisy to Native: LLM-driven Graph Restoration for Test-Time Graph Domain Adaptation
• An approach for systematic decomposition of complex llm tasks
• GCPO: When Contrast Fails, Go Gold
• Strategic Communication under Threat: Learning Information Trade-offs in Pursuit-Evasion Games
• An LLM-Powered Cooperative Framework for Large-Scale Multi-Vehicle Navigation
• FinMR: A Knowledge-Intensive Multimodal Benchmark for Advanced Financial Reasoning
• Augur: Modeling Covariate Causal Associations in Time Series via Large Language Models
• Understanding DeepResearch via Reports
• Towards Meaningful Transparency in Civic AI Systems
• Profit Mirage: Revisiting Information Leakage in LLM-based Financial Agents
• Enabling Personalized Long-term Interactions in LLM-based Agents through Persistent Memory and User Profiles
• Agent-Based Genetic Algorithm for Crypto Trading Strategy Optimization
• TaoSR-SHE: Stepwise Hybrid Examination Reinforcement Learning Framework for E-commerce Search Relevance
• VoiceAgentBench: Are Voice Assistants ready for agentic tasks?
• ReInAgent: A Context-Aware GUI Agent Enabling Human-in-the-Loop Mobile Task Navigation
• Language Models Do Not Embed Numbers Continuously
• PEAR: Phase Entropy Aware Reward for Efficient Reasoning
• AILoRA: Function-Aware Asymmetric Initialization for Low-Rank Adaptation of Large Language Models
• LinguaSim: Interactive Multi-Vehicle Testing Scenario Generation via Natural Language Instruction Based on Large Language Models
• Multi-Condition Conformal Selection
• AutoQual: An LLM Agent for Automated Discovery of Interpretable Features for Review Quality Assessment
• From Ethical Declarations to Provable Independence: An Ontology-Driven Optimal-Transport Framework for Certifiably Fair AI Systems
• Can Risk-taking AI-Assistants suitably represent entities
• Prepared mind, fast response: A temporal decoupling framework for adaptive knowledge orchestration in open-domain dialogue
• R-Horizon: How Far Can Your Large Reasoning Model Really Go in Breadth and Depth?
• Measuring What Matters: The AI Pluralism Index
• The Tournament Tree Method for preference elicitation in Multi-criteria decision-making
• DODO: Causal Structure Learning with Budgeted Interventions
• Selection, Reflection and Self-Refinement: Revisit Reasoning Tasks via a Causal Lens
• Chain-of-Trigger: An Agentic Backdoor that Paradoxically Enhances Agentic Robustness
• Co-TAP: Three-Layer Agent Interaction Protocol Technical Report
• Symmetry-Aware Fully-Amortized Optimization with Scale Equivariant Graph Metanetworks
• First Try Matters: Revisiting the Role of Reflection in Reasoning Models
• Beyond Pass@k: Breadth-Depth Metrics for Reasoning Boundaries
• LLMs Reproduce Human Purchase Intent via Semantic Similarity Elicitation of Likert Ratings
• QAgent: A modular Search Agent with Interactive Query Understanding
• Revisiting Hallucination Detection with Effective Rank-based Uncertainty
• Looking to Learn: Token-wise Dynamic Gating for Low-Resource Vision-Language Modelling
• AutoMLGen: Navigating Fine-Grained Optimization for Coding Agents
• CaRT: Teaching LLM Agents to Know When They Know Enough
• FlowSearch: Advancing deep research with dynamic structured knowledge flow
• Agent Learning via Early Experience
• How to Teach Large Multimodal Models New Skills
• Deep Learning Based Approach to Enhanced Recognition of Emotions and Behavioral Patterns of Autistic Children
• MultiFair: Multimodal Balanced Fairness-Aware Medical Classification with Dual-Level Gradient Modulation
• Local MAP Sampling for Diffusion Models
• Mitigating Surgical Data Imbalance with Dual-Prediction Video Diffusion Model
• Encode, Think, Decode: Scaling test-time reasoning with recursive latent thoughts
• Attention to Order: Transformers Discover Phase Transitions via Learnability
• Quantum Grid Path Planning Using Parallel QAOA Circuits Based on Minimum Energy Principle
• Haystack Engineering: Context Engineering for Heterogeneous and Agentic Long-Context Evaluation
• LASER: An LLM-based ASR Scoring and Evaluation Rubric
• Minimizing the Value-at-Risk of Loan Portfolio via Deep Neural Networks
• MoGU: Mixture-of-Gaussians with Uncertainty-based Gating for Time Series Forecasting
• HEMERA: A Human-Explainable Transformer Model for Estimating Lung Cancer Risk using GWAS Data
• Can Lessons From Human Teams Be Applied to Multi-Agent Systems? The Role of Structure, Diversity, and Interaction Dynamics
• A Denoising Framework for Real-World Ultra-Low Dose Lung CT Images Based on an Image Purification Strategy
• Can Speech LLMs Think while Listening?
• When Thoughts Meet Facts: Reusable Reasoning for Long-Context LMs
• MLLM4TS: Leveraging Vision and Multimodal Language Models for General Time-Series Analysis
• EEG Sleep Stage Classification with Continuous Wavelet Transform and Deep Learning
• OWL: Overcoming Window Length-Dependence in Speculative Decoding for Long-Context Inputs
• TRAVL: A Recipe for Making Video-Language Models Better Judges of Physics Implausibility
• Label Semantics for Robust Hyperspectral Image Classification
• Investigating Thematic Patterns and User Preferences in LLM Interactions using BERTopic
• Multi-Task Pre-Finetuning of Lightweight Transformer Encoders for Text Classification and NER
• Accuracy, Memory Efficiency and Generalization: A Comparative Study on Liquid Neural Networks and Recurrent Neural Networks
• Linguistic Patterns in Pandemic-Related Content: A Comparative Analysis of COVID-19, Constraint, and Monkeypox Datasets
• TGM: a Modular and Efficient Library for Machine Learning on Temporal Graphs
• Vocabulary embeddings organize linguistic structure early in language model training
• DGTEN: A Robust Deep Gaussian based Graph Neural Network for Dynamic Trust Evaluation with Uncertainty-Quantification Support
• Retentive Relevance: Capturing Long-Term User Value in Recommendation Systems
• Banking Done Right: Redefining Retail Banking with Language-Centric AI
• Value Flows
• OBCache: Optimal Brain KV Cache Pruning for Efficient Long-Context LLM Inference
• IKNet: Interpretable Stock Price Prediction via Keyword-Guided Integration of News and Technical Indicators
• TCIP: Threshold-Controlled Iterative Pyramid Network for Deformable Medical Image Registration
• Controllable Video Synthesis via Variational Inference
• Curriculum Learning with Synthetic Data for Enhanced Pulmonary Nodule Detection in Chest Radiographs
• Stress-Testing Model Specs Reveals Character Differences among Language Models
• Rethinking Reasoning: A Survey on Reasoning-based Backdoors in LLMs
• Causality Guided Representation Learning for Cross-Style Hate Speech Detection
• DEAS: DEtached value learning with Action Sequence for Scalable Offline RL
• MeSH: Memory-as-State-Highways for Recursive Transformers
• AppForge: From Assistant to Independent Developer - Are GPTs Ready for Software Development?
• UltraLED: Learning to See Everything in Ultra-High Dynamic Range Scenes
• Parallel Test-Time Scaling for Latent Reasoning Models
• A Unified Multi-Task Learning Framework for Generative Auto-Bidding with Validation-Aligned Optimization
• ToolLibGen: Scalable Automatic Tool Creation and Aggregation for LLM Reasoning