AI Research News Feeds for October 1st, 2025

AI RESEARCH PAPERS & ACADEMIC SOURCES

• From Ideal to Real: Unified and Data-Efficient Dense Prediction for Real-World Scenarios
• Learning an Ensemble Token from Task-driven Priors in Facial Analysis
• Taming Diffusion Transformer for Efficient Mobile Video Generation in Seconds
• tCURLoRA: Tensor CUR Decomposition Based Low-Rank Parameter Adaptation and Its Application in Medical Image Segmentation
• Towards autonomous photogrammetric forest inventory using a lightweight under-canopy robotic drone
• PolSAM: Polarimetric Scattering Mechanism Informed Segment Anything Model
• J-NeuS: Joint field optimization for Neural Surface reconstruction in urban scenes with limited image overlap
• Binary Diffusion Probabilistic Model
• RobustMerge: Parameter-Efficient Model Merging for MLLMs with Direction Robustness
• UFO: A Unified Approach to Fine-grained Visual Perception via Open-ended Language Interface
• DecAlign: Hierarchical Cross-Modal Alignment for Decoupled Multimodal Representation Learning
• A Large Scale Analysis of Gender Biases in Text-to-Image Generative Models
• BoundMatch: Boundary detection applied to semi-supervised segmentation
• SuperEvent: Cross-Modal Learning of Event-based Keypoint Detection for SLAM
• BoxDreamer: Dreaming Box Corners for Generalizable Object Pose Estimation
• Neural Catalog: Scaling Species Recognition with Catalog of Life-Augmented Generation
• KDC-Diff: A Latent-Aware Diffusion Model with Knowledge Retention for Memory-Efficient Image Generation
• StPR: Spatiotemporal Preservation and Routing for Exemplar-Free Video Class-Incremental Learning
• Multi-View Projection for Unsupervised Domain Adaptation in 3D Semantic Segmentation
• How Do Large Vision-Language Models See Text in Image? Unveiling the Distinctive Role of OCR Heads
• Photography Perspective Composition: Towards Aesthetic Perspective Recommendation
• AVCD: Mitigating Hallucinations in Audio-Visual Large Language Models through Contrastive Decoding
• Fork-Merge Decoding: Enhancing Multimodal Understanding in Audio-Visual Large Language Models
• Anatomy-DT: A Cross-Diffusion Digital Twin for Anatomical Evolution
• Online Mapping for Autonomous Driving: Addressing Sensor Generalization and Dynamic Map Updates in Campus Environments
• LTA-L2S: Lexical Tone-Aware Lip-to-Speech Synthesis for Mandarin with Cross-Lingual Transfer Learning
• dVLA: Diffusion Vision-Language-Action Model with Multimodal Chain-of-Thought
• Multi-modal Liver Segmentation and Fibrosis Staging Using Real-world MRI Images
• MR$^2$-Bench: Going Beyond Matching to Reasoning in Multimodal Retrieval
• GastroViT: A Vision Transformer Based Ensemble Learning Approach for Gastrointestinal Disease Classification with Grad CAM & SHAP Visualization
• Automated and Scalable SEM Image Analysis of Perovskite Solar Cell Materials via a Deep Segmentation Framework
• M$^{2}$SNet: Multi-scale in Multi-scale Subtraction Network for Medical Image Segmentation
• RealLiFe: Real-Time Light Field Reconstruction via Hierarchical Sparse Gradient Descent
• Developing Generalist Foundation Models from a Multimodal Dataset for 3D Computed Tomography
• Adaptive Modality Balanced Online Knowledge Distillation for Brain-Eye-Computer based Dim Object Detection
• LoRA-PT: Low-Rank Adapting UNETR for Hippocampus Segmentation Using Principal Tensor Singular Values and Vectors
• Rethinking Weak-to-Strong Augmentation in Source-Free Domain Adaptive Object Detection
• Multi-temporal crack segmentation in concrete structures using deep learning approaches
• ResGS: Residual Densification of 3D Gaussian for Efficient Detail Recovery
• PRISM: Progressive Rain removal with Integrated State-space Modeling
• Post-Training Quantization via Residual Truncation and Zero Suppression for Diffusion Models
• Multi-View Camera System for Variant-Aware Autonomous Vehicle Inspection and Defect Detection
• Stylos: Multi-View 3D Stylization with Single-Forward Gaussian Splatting
• CBAM Integrated Attention Driven Model For Betel Leaf Diseases Classification With Explainable AI
• Revealing the Power of Post-Training for Small Language Models via Knowledge Distillation
• DEPTHOR++: Robust Depth Enhancement from a Real-World Lightweight dToF and RGB Guidance
• Stable Cinemetrics : Structured Taxonomy and Evaluation for Professional Video Generation
• Autoproof: Automated Segmentation Proofreading for Connectomics
• DiffCamera: Arbitrary Refocusing on Images
• Video Object Segmentation-Aware Audio Generation
• Hy-Facial: Hybrid Feature Extraction by Dimensionality Reduction Methods for Enhanced Facial Expression Classification
• DA$^2$: Depth Anything in Any Direction
• HART: Human Aligned Reconstruction Transformer
• Benchmarking Egocentric Visual-Inertial SLAM at City Scale
• Query-Kontext: An Unified Multimodal Model for Image Generation and Editing
• TTT3R: 3D Reconstruction as Test-Time Training
• Challenges and Solutions in Selecting Optimal Lossless Data Compression Algorithms
• Geometric Learning of Canonical Parameterizations of $2D$-curves
• EasyOcc: 3D Pseudo-Label Supervision for Fully Self-Supervised Semantic Occupancy Prediction Models
• Predicting Penalty Kick Direction Using Multi-Modal Deep Learning with Pose-Guided Attention
• EchoGen: Generating Visual Echoes in Any Scene via Feed-Forward Subject-Driven Auto-Regressive Model
• Human-MME: A Holistic Evaluation Benchmark for Human-Centric Multimodal Large Language Models
• Beyond Overall Accuracy: Pose- and Occlusion-driven Fairness Analysis in Pedestrian Detection for Autonomous Driving
• TSalV360: A Method and Dataset for Text-driven Saliency Detection in 360-Degrees Videos
• Generalized Fine-Grained Category Discovery with Multi-Granularity Conceptual Experts
• IMG: Calibrating Diffusion Models via Implicit Multimodal Guidance
• Interpret, prune and distill Donut : towards lightweight VLMs for VQA on document
• Seeing Space and Motion: Enhancing Latent Actions with Spatial and Dynamic Awareness for VLA
• Cat: Post-training quantization error reduction via cluster-based affine transformation
• Continuous Space-Time Video Super-Resolution with 3D Fourier Fields
• SQUARE: Semantic Query-Augmented Fusion and Efficient Batch Reranking for Training-free Zero-Shot Composed Image Retrieval
• Go with Your Gut: Scaling Confidence for Autoregressive Image Generation
• PANDA: Towards Generalist Video Anomaly Detection via Agentic AI Engineer
• MotionRAG: Motion Retrieval-Augmented Image-to-Video Generation
• Image-Difficulty-Aware Evaluation of Super-Resolution Models
• LiDAR Point Cloud Colourisation Using Multi-Camera Fusion and Low-Light Image Enhancement
• MAPLE: Multi-scale Attribute-enhanced Prompt Learning for Few-shot Whole Slide Image Classification
• DeepSketcher: Internalizing Visual Manipulation for Multimodal Reasoning
• LLaVAShield: Safeguarding Multimodal Multi-Turn Dialogues in Vision-Language Models
• UniMMAD: Unified Multi-Modal and Multi-Class Anomaly Detection via MoE-Driven Feature Decompression
• Self-Supervised Anatomical Consistency Learning for Vision-Grounded Medical Report Generation
• A Multi-purpose Tracking Framework for Salmon Welfare Monitoring in Challenging Environments
• PinPoint3D: Fine-Grained 3D Part Segmentation from a Few Clicks
• Towards Reliable and Holistic Visual In-Context Learning Prompt Selection
• AgenticIQA: An Agentic Framework for Adaptive and Interpretable Image Quality Assessment
• New Fourth-Order Grayscale Indicator-Based Telegraph Diffusion Model for Image Despeckling
• SETR: A Two-Stage Semantic-Enhanced Framework for Zero-Shot Composed Image Retrieval
• GeoLink: Empowering Remote Sensing Foundation Model with OpenStreetMap Data
• PatchVSR: Breaking Video Diffusion Resolution Limits with Patch-wise Video Super-Resolution
• Causally Guided Gaussian Perturbations for Out-Of-Distribution Generalization in Medical Imaging
• SGS: Segmentation-Guided Scoring for Global Scene Inconsistencies
• DGM4+: Dataset Extension for Global Scene Inconsistency
• AIMCoT: Active Information-driven Multimodal Chain-of-Thought for Vision-Language Reasoning
• How Diffusion Models Memorize
• ProbMed: A Probabilistic Framework for Medical Multimodal Binding
• SAGE: Spatial-visual Adaptive Graph Exploration for Visual Place Recognition
• LaTo: Landmark-tokenized Diffusion Transformer for Fine-grained Human Face Editing
• The 1st Solution for MOSEv1 Challenge on LSVOS 2025: CGFSeg
• LieHMR: Autoregressive Human Mesh Recovery with $SO(3)$ Diffusion
• Dragging with Geometry: From Pixels to Geometry-Guided Image Editing
• IPDRecon: Image-Plane Geometric Decoding for View-Invariant Indoor Scene Reconstruction
• ART-VITON: Measurement-Guided Latent Diffusion for Artifact-Free Virtual Try-On
• PCPO: Proportionate Credit Policy Optimization for Aligning Image Generation Models
• Self-Evolving Vision-Language Models for Image Quality Assessment via Voting and Ranking
• EchoingECG: An Electrocardiogram Cross-Modal Model for Echocardiogram Tasks
• Adapting SAM with Dynamic Similarity Graphs for Few-Shot Parameter-Efficient Small Dense Object Detection: A Case Study of Chickpea Pods in Field Conditions
• Overview of GeoLifeCLEF 2023: Species Composition Prediction with High Spatial Resolution at Continental Scale Using Remote Sensing
• MuSLR: Multimodal Symbolic Logical Reasoning
• PatchEAD: Unifying Industrial Visual Prompting Frameworks for Patch-Exclusive Anomaly Detection
• Seeing Through Deception: Uncovering Misleading Creator Intent in Multimodal News with Vision-Language Models
• ReLoop: "Seeing Twice and Thinking Backwards" via Closed-loop Training to Mitigate Hallucinations in Multimodal understanding
• Taming the Tri-Space Tension: ARC-Guided Hallucination Modeling and Control for Text-to-Image Generation
• LUMA: Low-Dimension Unified Motion Alignment with Dual-Path Anchoring for Text-to-Motion Diffusion Model
• Editing Physiological Signals in Videos Using Latent Representations
• DepthLM: Metric Depth From Vision Language Models
• Infrastructure Sensor-enabled Vehicle Data Generation using Multi-Sensor Fusion for Proactive Safety Applications at Work Zone
• Seeing Before Reasoning: A Unified Framework for Generalizable and Explainable Fake Image Detection
• Robust Visual Localization in Compute-Constrained Environments by Salient Edge Rendering and Weighted Hamming Similarity
• FishNet++: Analyzing the capabilities of Multimodal Large Language Models in marine biology
• GaussianLens: Localized High-Resolution Reconstruction via On-Demand Gaussian Densification
• LMOD+: A Comprehensive Multimodal Dataset and Benchmark for Developing and Evaluating Multimodal Large Language Models in Ophthalmology
• Anchor-free Cross-view Object Geo-localization with Gaussian Position Encoding and Cross-view Association
• DescribeEarth: Describe Anything for Remote Sensing Images
• OmniDFA: A Unified Framework for Open Set Synthesis Image Detection and Few-Shot Attribution
• Cut the Deadwood Out: Backdoor Purification via Guided Module Substitution
• SAFE-SQL: Self-Augmented In-Context Learning with Fine-grained Example Selection for Text-to-SQL
• Safety is Not Only About Refusal: Reasoning-Enhanced Fine-tuning for Interpretable LLM Safety
• A Position Paper on the Automatic Generation of Machine Learning Leaderboards
• Wolf Hidden in Sheep's Conversations: Toward Harmless Data-Based Backdoor Attacks for Jailbreaking Large Language Models
• Frankentext: Stitching random text fragments into long-form narratives
• v1: Learning to Point Visual Tokens for Multimodal Grounded Reasoning
• ConfRAG: Confidence-Guided Retrieval-Augmenting Generation
• TableRAG: A Retrieval Augmented Generation Framework for Heterogeneous Document Reasoning
• Medical Red Teaming Protocol of Language Models: On the Importance of User Perspectives in Healthcare Settings
• Text-Based Approaches to Item Alignment to Content Standards in Large-Scale Reading & Writing Tests
• CreAgentive: An Agent Workflow Driven Multi-Category Creative Generation Engine
• dParallel: Learnable Parallel Decoding for dLLMs
• BatonVoice: An Operationalist Framework for Enhancing Controllable Speech Synthesis with Linguistic Intelligence from LLMs
• Training Matryoshka Mixture-of-Experts for Elastic Inference-Time Expert Utilization
• Towards Reliable Benchmarking: A Contamination Free, Controllable Evaluation Framework for Multi-step LLM Function Calling
• Generating Difficult-to-Translate Texts
• Scaling Spoken Language Models with Syllabic Speech Tokenization
• ActorDB: A Unified Database Model Integrating Single-Writer Actors, Incremental View Maintenance, and Zero-Trust Messaging
• Fingerprinting LLMs via Prompt Injection
• FinCap: Topic-Aligned Captions for Short-Form Financial YouTube Videos
• A Multimodal LLM Approach for Visual Question Answering on Multiparametric 3D Brain MRI
• VLM-FO1: Bridging the Gap Between High-Level Reasoning and Fine-Grained Perception in VLMs
• ProfVLM: A Lightweight Video-Language Model for Multi-View Proficiency Estimation
• Exploring the Impact of Data Quantity on ASR in Extremely Low-resource Languages
• Aristotle: Mastering Logical Reasoning with A Logic-Complete Decompose-Search-Resolve Framework
• Vocabulary Customization for Efficient Domain-Specific LLM Deployment
• The Hunger Game Debate: On the Emergence of Over-Competition in Multi-Agent Systems
• CliniBench: A Clinical Outcome Prediction Benchmark for Generative and Encoder-Based Language Models
• MGen: Millions of Naturally Occurring Generics in Context
• Explaining novel senses using definition generation with open language models
• VietBinoculars: A Zero-Shot Approach for Detecting Vietnamese LLM-Generated Text
• Optimizing Speech Language Models for Acoustic Consistency
• One-Token Rollout: Guiding Supervised Fine-Tuning of LLMs with Policy Gradient
• Latent Thinking Optimization: Your Latent Reasoning Language Model Secretly Encodes Reward Signals in its Latent Thoughts
• Fast-dLLM v2: Efficient Block-Diffusion LLM
• An Annotation Scheme for Factuality and its Application to Parliamentary Proceedings
• Automatic Fact-checking in English and Telugu
• DyFlow: Dynamic Workflow Framework for Agentic Reasoning
• The Silent Judge: Unacknowledged Shortcut Bias in LLM-as-a-Judge
• Limited Preference Data? Learning Better Reward Model with Latent Space Synthesis
• IMProofBench: Benchmarking AI on Research-Level Mathematical Proof Generation
• Reinforced Strategy Optimization for Conversational Recommender Systems via Network-of-Experts
• The Flaw of Averages: Quantifying Uniformity of Performance on Benchmarks
• Mitigating Biases in Language Models via Bias Unlearning
• Atomic Thinking of LLMs: Decoupling and Exploring Mathematical Reasoning Abilities
• CATCH: A Novel Data Synthesis Framework for High Therapy Fidelity and Memory-Driven Planning Chain of Thought in AI Counseling
• Assessing Algorithmic Bias in Language-Based Depression Detection: A Comparison of DNN and LLM Approaches
• ReTAG: Retrieval-Enhanced, Topic-Augmented Graph-Based Global Sensemaking
• Personalized Scientific Figure Caption Generation: An Empirical Study on Author-Specific Writing Style Transfer
• Believing without Seeing: Quality Scores for Contextualizing Vision-Language Model Explanations
• ReFACT: A Benchmark for Scientific Confabulation Detection with Positional Error Annotations
• ASR Under Noise: Exploring Robustness for Sundanese and Javanese
• Mem-{\alpha}: Learning Memory Construction via Reinforcement Learning
• Understanding the Mixture-of-Experts with Nadaraya-Watson Kernel
• Bringing Emerging Architectures to Sequence Labeling in NLP
• Reliability Crisis of Reference-free Metrics for Grammatical Error Correction
• RAGferee: Building Contextual Reward Models for Retrieval-Augmented Generation
• RE$^2$: Improving Chinese Grammatical Error Correction via Retrieving Appropriate Examples with Explanation
• Unspoken Hints: Accuracy Without Acknowledgement in LLM Reasoning
• RE-Searcher: Robust Agentic Search with Goal-oriented Planning and Self-reflection
• The Rise of AfricaNLP: Contributions, Contributors, and Community Impact (2005-2025)
• Beyond WER: Probing Whisper's Sub-token Decoder Across Diverse Language Resource Levels
• Performance and competence intertwined: A computational model of the Null Subject stage in English-speaking children
• Don't Sweat the Small Stuff: Segment-Level Meta-Evaluation Based on Pairwise Difference Correlation
• Transformers through the lens of support-preserving maps between measures
• The Media Bias Detector: A Framework for Annotating and Analyzing the News at Scale
• QFrBLiMP: a Quebec-French Benchmark of Linguistic Minimal Pairs
• Apple: Toward General Active Perception via Reinforcement Learning
• A quantitative analysis of semantic information in deep representations of text and images
• GIM: Improved Interpretability for Large Language Models
• Discovering and Steering Interpretable Concepts in Large Generative Music Models
• tenSVD algorithm for compression
• A Copula Based Supervised Filter for Feature Selection in Diabetes Risk Prediction Using Machine Learning
• KVzip: Query-Agnostic KV Cache Compression with Context Reconstruction
• LoLA: Low-Rank Linear Attention With Sparse Caching
• Regularizing Learnable Feature Extraction for Automatic Speech Recognition
• TADA: Improved Diffusion Sampling with Training-free Augmented Dynamics
• Regret Analysis of Posterior Sampling-Based Expected Improvement for Bayesian Optimization
• BEDTime: A Unified Benchmark for Automatically Describing Time Series
• Asymptotic Classification Error for Heavy-Tailed Renewal Processes
• Efficient Fairness-Performance Pareto Front Computation
• Brain Tumor Classification on MRI in Light of Molecular Markers
• Information Design with Unknown Prior
• Unified Cross-Modal Image Synthesis with Hierarchical Mixture of Product-of-Experts
• Watermark under Fire: A Robustness Evaluation of LLM Watermarking
• LargeAD: Large-Scale Cross-Sensor Data Pretraining for Autonomous Driving
• Complexity Analysis of Normalizing Constant Estimation: from Jarzynski Equality to Annealed Importance Sampling and beyond
• Surrogate models for diffusion on graphs via sparse polynomials
• Scalable Fingerprinting of Large Language Models
• Controllable Motion Generation via Diffusion Modal Coupling
• A Review on Riemannian Metric Learning: Closer to You than You Imagine
• Approximation properties of neural ODEs
• ThinkEdit: Interpretable Weight Editing to Mitigate Overly Short Thinking in Reasoning Models
• Neural Kinematic Bases for Fluids
• AutoJudge: Judge Decoding Without Manual Annotation
• Fast Likelihood-Free Parameter Estimation for L\'evy Processes
• Using Knowledge Graphs to harvest datasets for efficient CLIP model training
• Detecting Instruction Fine-tuning Attacks on Language Models using Influence Function
• Neural Multivariate Regression: Qualitative Insights from the Unconstrained Feature Model
• AutoRAN: Automated Hijacking of Safety Reasoning in Large Reasoning Models
• What Can RL Bring to VLA Generalization? An Empirical Study
• Feature-aware Hypergraph Generation via Next-Scale Prediction
• On Fitting Flow Models with Large Sinkhorn Couplings
• Flatness After All?
• IMPACT: Importance-Aware Activation Space Reconstruction
• Towards Robust Surrogate Models: Benchmarking Machine Learning Approaches to Expediting Phase Field Simulations of Brittle Fracture
• The Serial Scaling Hypothesis
• FlowCast-ODE: Continuous Hourly Weather Forecasting with Dynamic Flow Matching and ODE Solver
• Bengali Document Layout Analysis -- A YOLOV8 Based Ensembling Approach
• Minimum Description Feature Selection for Complexity Reduction in Machine Learning-based Wireless Positioning
• Fast training of accurate physics-informed neural networks without gradient descent
• RobustNeuralNetworks.jl: a Package for Machine Learning and Data-Driven Control with Certified Robustness
• Efficiently Escaping Saddle Points for Policy Optimization
• Model Extraction Attacks Revisited
• Collective Counterfactual Explanations: Balancing Individual Goals and Collective Dynamics
• Complexity Reduction in Machine Learning-Based Wireless Positioning: Minimum Description Features
• From Mean to Extreme: Formal Differential Privacy Bounds on the Success of Real-World Data Reconstruction Attacks
• FedGCS: A Generative Framework for Efficient Client Selection in Federated Learning via Gradient-based Optimization
• Hardware-Aware Parallel Prompt Decoding for Memory-Efficient Acceleration of LLM Inference
• Amelia: A Large Dataset and Model for Airport Surface Movement Forecasting
• Large-Scale Targeted Cause Discovery via Learning from Simulated Data
• Towards Convexity in Anomaly Detection: A New Formulation of SSLM with Unique Optimal Solutions
• Information-Geometric Barycenters for Bayesian Federated Learning
• Learning Theory for Kernel Bilevel Optimization
• Teaching Metric Distance to Discrete Autoregressive Language Models
• InverseBench: Benchmarking Plug-and-Play Diffusion Priors for Inverse Problems in Physical Sciences
• Understanding Formal Reasoning Failures in LLMs as Abstract Interpreters
• Beyond Next Token Probabilities: Learnable, Fast Detection of Hallucinations and Data Contamination on LLM Output Distributions
• Identifying and Evaluating Inactive Heads in Pretrained LLMs
• TrackCore-F: Deploying Transformer-Based Subatomic Particle Tracking on FPGAs
• Are neural scaling laws leading quantum chemistry astray?
• TrackFormers Part 2: Enhanced Transformer-Based Models for High-Energy Physics Track Reconstruction
• An Orthogonal Learner for Individualized Outcomes in Markov Decision Processes
• Stabilization of nonlinear systems with unknown delays via delay-adaptive neural operator approximate predictors
• Contrastive Diffusion Guidance for Spatial Inverse Problems
• Signal-Aware Workload Shifting Algorithms with Uncertainty-Quantified Predictors
• Ferret-UI Lite: Lessons from Building Small On-Device GUI Agents
• Towards Verified Code Reasoning by LLMs
• Pretrain-Test Task Alignment Governs Generalization in In-Context Learning
• Estimating Dimensionality of Neural Representations from Finite Samples
• DeepProv: Behavioral Characterization and Repair of Neural Networks via Inference Provenance Graph Analysis
• Source Separation for A Cappella Music
• DeepScientist: Advancing Frontier-Pushing Scientific Findings Progressively
• Convergence and Divergence of Language Models under Different Random Seeds
• Max-Sliced Wasserstein Distance and its use for GANs
• CO3: Contrasting Concepts Compose Better
• Scaling Equilibrium Propagation to Deeper Neural Network Architectures
• BALLAST: Bayesian Active Learning with Look-ahead Amendment for Sea-drifter Trajectories under Spatio-Temporal Vector Fields
• GaussEdit: Adaptive 3D Scene Editing with Text and Image Prompts
• Text-to-Scene with Large Reasoning Models
• Efficient Distributed Training via Dual Batch Sizes and Cyclic Progressive Learning
• EVODiff: Entropy-aware Variance Optimized Diffusion Inference
• Ordinal Label-Distribution Learning with Constrained Asymmetric Priors for Imbalanced Retinal Grading
• Non-Vacuous Generalization Bounds: Can Rescaling Invariances Help?
• Benchmarking Diarization Models
• Self-supervised learning for phase retrieval
• The silence of the weights: an investigation of structural pruning strategies for attention-based audio signal architectures
• Hybrid Quantum-Classical Optimisation of Traveling Salesperson Problem
• Why is topology hard to learn?
• PRPO: Paragraph-level Policy Optimization for Vision-Language Deepfake Detection
• FLOWER: A Flow-Matching Solver for Inverse Problems
• Ultra-Reliable Risk-Aggregated Sum Rate Maximization via Model-Aided Deep Learning
• TAU: A Benchmark for Cultural Sound Understanding Beyond Semantics
• Personalized Auto-Grading and Feedback System for Constructive Geometry Tasks Using Large Language Models on an Online Math Platform
• Enhancing Split Learning with Sharded and Blockchain-Enabled SplitFed Approaches
• Conservative Decisions with Risk Scores
• MetaChest: Generalized few-shot learning of patologies from chest X-rays
• Coupling Generative Modeling and an Autoencoder with the Causal Bridge
• RFG: Test-Time Scaling for Diffusion Large Language Model Reasoning with Reward-Free Guidance
• When Langevin Monte Carlo Meets Randomization: Non-asymptotic Error Bounds beyond Log-Concavity and Gradient Lipschitzness
• Generalized Contrastive Learning for Universal Multimodal Retrieval
• Using Images from a Video Game to Improve the Detection of Truck Axles
• Importance Sampling for Multi-Negative Multimodal Direct Preference Optimization
• Transformer-Based Rate Prediction for Multi-Band Cellular Handsets
• Test time training enhances in-context learning of nonlinear functions
• Detecting Hope Across Languages: Multiclass Classification for Positive Online Discourse
• SAC Flow: Sample-Efficient Reinforcement Learning of Flow-Based Policies via Velocity-Reparameterized Sequential Modeling
• Sharpness of Minima in Deep Matrix Factorization: Exact Expressions
• Logo-VGR: Visual Grounded Reasoning for Open-world Logo Recognition
• RoBiologyDataChoiceQA: A Romanian Dataset for improving Biology understanding of Large Language Models
• Better Privilege Separation for Agents by Restricting Data Types
• AccidentBench: Benchmarking Multimodal Understanding and Reasoning in Vehicle Accidents and Beyond
• SPATA: Systematic Pattern Analysis for Detailed and Transparent Data Cards
• Understanding Practitioners Perspectives on Monitoring Machine Learning Systems
• Cyclic Ablation: Testing Concept Localization against Functional Regeneration in AI
• RANGER -- Repository-Level Agent for Graph-Enhanced Retrieval
• Evaluating the Impact of Radiographic Noise on Chest X-ray Semantic Segmentation and Disease Classification Using a Scalable Noise Injection Framework
• Position-Blind Ptychography: Viability of image reconstruction via data-driven variational inference
• Mechanisms of Matter: Language Inferential Benchmark on Physicochemical Hypothesis in Materials Synthesis
• Aspects of holographic entanglement using physics-informed-neural-networks
• Bayesian Transformer for Pan-Arctic Sea Ice Concentration Mapping and Uncertainty Estimation using Sentinel-1, RCM, and AMSR2 Data
• Neural Optimal Transport Meets Multivariate Conformal Prediction
• SimulRAG: Simulator-based RAG for Grounding LLMs in Long-form Scientific QA
• Fair Classification by Direct Intervention on Operating Characteristics
• Scalable Boltzmann Generators for equilibrium sampling of large-scale materials
• One-shot Conditional Sampling: MMD meets Nearest Neighbors
• AGNOMIN - Architecture Agnostic Multi-Label Function Name Prediction
• Defeating Cerberus: Concept-Guided Privacy-Leakage Mitigation in Multimodal Language Models
• LLM-Assisted Emergency Triage Benchmark: Bridging Hospital-Rich and MCI-Like Field Simulation
• Data-to-Energy Stochastic Dynamics
• Refine Drugs, Don't Complete Them: Uniform-Source Discrete Flows for Fragment-Based Drug Discovery
• Extensions of Robbins-Siegmund Theorem with Applications in Reinforcement Learning
• fev-bench: A Realistic Benchmark for Time Series Forecasting
• DiVeQ: Differentiable Vector Quantization Using the Reparameterization Trick
• Equivariance by Local Canonicalization: A Matter of Representation
• Entropy After $\langle \texttt{/Think} \rangle$ for reasoning model early exiting
• Machine-Learning Driven Load Shedding to Mitigate Instability Attacks in Power Grids
• The Loss Kernel: A Geometric Probe for Deep Learning Interpretability
• TASP: Topology-aware Sequence Parallelism
• Bayesian Influence Functions for Hessian-Free Data Attribution
• Importance of localized dilatation and distensibility in identifying determinants of thoracic aortic aneurysm with neural operators
• Linking Process to Outcome: Conditional Reward Modeling for LLM Reasoning
• Clarification as Supervision: Reinforcement Learning for Vision-Language Interfaces
• Uncertainty Quantification for Regression using Proper Scoring Rules
• Recursive Self-Aggregation Unlocks Deep Thinking in Large Language Models
• Attention as a Compass: Efficient Exploration for Process-Supervised RL in Reasoning Models
• Alignment-Aware Decoding
• Neighbor-aware informal settlement mapping with graph convolutional networks
• PDE Solvers Should Be Local: Fast, Stable Rollouts with Learned Local Stencils
• Marginal Flow: a flexible and efficient framework for density estimation
• Thinking-Free Policy Initialization Makes Distilled Reasoning Models More Effective and Efficient Reasoners
• Machine Learning Detection of Lithium Plating in Lithium-ion Cells: A Gaussian Process Approach
• Beyond Linear Probes: Dynamic Safety Monitoring for Language Models
• From Fragile to Certified: Wasserstein Audits of Group Fairness Under Distribution Shift
• Wasserstein Distributionally Robust Optimization Through the Lens of Structural Causal Models and Individual Fairness
• Reframing Generative Models for Physical Systems using Stochastic Interpolants
• Tuning the Tuner: Introducing Hyperparameter Optimization for Auto-Tuning
• NeuroTTT: Bridging Pretraining-Downstream Task Misalignment in EEG Foundation Models via Test-Time Training
• Attribution-Guided Decoding
• A Review on Single-Problem Multi-Attempt Heuristic Optimization
• ACE: Adapting sampling for Counterfactual Explanations
• A Generalized Information Bottleneck Theory of Deep Learning
• FedMuon: Federated Learning with Bias-corrected LMO-based Optimization
• Memory-Driven Self-Improvement for Decision Making with Large Language Models
• Improving Sampling Efficiency in RLVR through Adaptive Rollout and Response Reuse
• Decentralized Asynchronous Multi-player Bandits
• Kairos: Towards Adaptive and Generalizable Time Series Foundation Models
• MIDAS: Misalignment-based Data Augmentation Strategy for Imbalanced Multimodal Learning
• RL-Guided Data Selection for Language Model Finetuning
• Federated Learning with Enhanced Privacy via Model Splitting and Random Client Participation
• ReNF: Rethinking the Design Space of Neural Long-Term Time Series Forecasters
• Reevaluating Convolutional Neural Networks for Spectral Analysis: A Focus on Raman Spectroscopy
• Exact Solutions to the Quantum Schr\"odinger Bridge Problem
• CAST: Continuous and Differentiable Semi-Structured Sparsity-Aware Training for Large Language Models
• Informed Asymmetric Actor-Critic: Leveraging Privileged Signals Beyond Full-State Access
• FITS: Towards an AI-Driven Fashion Information Tool for Sustainability
• Stealthy Yet Effective: Distribution-Preserving Backdoor Attacks on Graph Classification
• Scaling Up Temporal Domain Generalization via Temporal Experts Averaging
• Clip-Low Increases Entropy and Clip-High Decreases Entropy in Reinforcement Learning of Large Language Models
• UncertainGen: Uncertainty-Aware Representations of DNA Sequences for Metagenomic Binning
• Domain-Aware Hyperdimensional Computing for Edge Smart Manufacturing
• Accelerating Transformers in Online RL
• Guiding Mixture-of-Experts with Temporal Multimodal Interactions
• Minimalist Explanation Generation and Circuit Discovery
• A Unified Probabilistic Framework for Dictionary Learning with Parsimonious Activation
• Can VLM Pseudo-Labels Train a Time-Series QA Model That Outperforms the VLM?
• Physics-Informed Learning for Human Whole-Body Kinematics Prediction via Sparse IMUs
• Adaptive Graph Coarsening for Efficient GNN Training
• Expert Merging: Model Merging with Unsupervised Expert Alignment and Importance-Guided Layer Chunking
• Reweighted Flow Matching via Unbalanced OT for Label-free Long-tailed Generation
• MuPlon: Multi-Path Causal Optimization for Claim Verification through Controlling Confounding
• Beyond Point Estimates: Likelihood-Based Full-Posterior Wireless Localization
• A Physics-Guided Probabilistic Surrogate Modeling Framework for Digital Twins of Underwater Radiated Noise
• Less is More: Towards Simple Graph Contrastive Learning
• Rotation Control Unlearning: Quantifying and Controlling Continuous Unlearning for LLM with The Cognitive Rotation Space
• OPPO: Accelerating PPO-based RLHF via Pipeline Overlap
• Online Decision Making with Generative Action Sets
• A Hamiltonian driven Geometric Construction of Neural Networks on the Lognormal Statistical Manifold
• From Cheap Geometry to Expensive Physics: Elevating Neural Operators via Latent Shape Pretraining
• Characterization and Learning of Causal Graphs with Latent Confounders and Post-treatment Selection from Interventional Data
• Scalable Disk-Based Approximate Nearest Neighbor Search with Page-Aligned Graph
• Can Molecular Foundation Models Know What They Don't Know? A Simple Remedy with Preference Optimization
• EEsizer: LLM-Based AI Agent for Sizing of Analog and Mixed Signal Circuit
• World Model for AI Autonomous Navigation in Mechanical Thrombectomy
• Flow Matching with Semidiscrete Couplings
• Meta-Router: Bridging Gold-standard and Preference-based Evaluations in Large Language Model Routing
• Lightweight and Robust Federated Data Valuation
• Safe In-Context Reinforcement Learning
• Machine Learning Algorithms for Improving Black Box Optimization Solvers
• Binary Sparse Coding for Interpretability
• Effective Model Pruning
• Layer-wise dynamic rank for compressing large language models
• Swift: An Autoregressive Consistency Model for Efficient Weather Forecasting
• How Does Preconditioning Guide Feature Learning in Deep Neural Networks?
• Deep set based operator learning with uncertainty quantification
• Growing Winning Subnetworks, Not Pruning Them: A Paradigm for Density Discovery in Sparse Neural Networks
• Nudging the Boundaries of LLM Reasoning
• Norm-Q: Effective Compression Method for Hidden Markov Models in Neuro-Symbolic Applications
• Conformal Prediction for Signal Temporal Logic Inference
• MSCoD: An Enhanced Bayesian Updating Framework with Multi-Scale Information Bottleneck and Cooperative Attention for Structure-Based Drug Design
• Integrated Forecasting of Marine Renewable Power: An Adaptively Bayesian-Optimized MVMD-LSTM Framework for Wind-Solar-Wave Energy
• Simple, Fast and Efficient Injective Manifold Density Estimation with Random Projections
• WDformer: A Wavelet-based Differential Transformer Model for Time Series Forecasting
• Sampling via Gaussian Mixture Approximations
• Fine-tuning of Large Language Models for Domain-Specific Cybersecurity Knowledge
• Heterogeneous Multi-agent Collaboration in UAV-assisted Mobile Crowdsensing Networks
• MAESTRO : Adaptive Sparse Attention and Robust Learning for Multimodal Dynamic Time Series
• Optimisation of Resource Allocation in Heterogeneous Wireless Networks Using Deep Reinforcement Learning
• Gradient Descent with Large Step Sizes: Chaos and Fractal Convergence Region
• Deep Survival Analysis for Competing Risk Modeling with Functional Covariates and Missing Data Imputation
• On the Shape of Latent Variables in a Denoising VAE-MoG: A Posterior Sampling-Based Study
• Crowdsourcing Without People: Modelling Clustering Algorithms as Experts
• Multi-Task Equation Discovery
• Leveraging Vulnerabilities in Temporal Graph Neural Networks via Strategic High-Impact Assaults
• Feedback Control for Small Budget Pacing
• SOLD: SELFIES-based Objective-driven Latent Diffusion
• VLHSA: Vision-Language Hierarchical Semantic Alignment for Jigsaw Puzzle Solving with Eroded Gaps
• Polynomial Contrastive Learning for Privacy-Preserving Representation Learning on Graphs
• Hyperbolic Optimization
• DPSformer: A long-tail-aware model for improving heavy rainfall prediction
• LEMs: A Primer On Large Execution Models
• Anomaly detection by partitioning of multi-variate time series
• Evaluating Double Descent in Machine Learning: Insights from Tree-Based Models Applied to a Genomic Prediction Task
• On The Dynamic Ensemble Selection for TinyML-based Systems -- a Preliminary Study
• Sensor optimization for urban wind estimation with cluster-based probabilistic framework
• AMLA: MUL by ADD in FlashAttention Rescaling
• Mind the Gap: A Review of Arabic Post-Training Datasets and Their Limitations
• The Impact of Language Mixing on Bilingual LLM Reasoning
• Measuring the Measures: Discriminative Capacity of Representational Similarity Metrics Across Model Families
• Long-Horizon Visual Imitation Learning via Plan and Code Reflection
• Multi Layered Autonomy and AI Ecologies in Robotic Art Installations
• Negative-Guided Subject Fidelity Optimization for Zero-Shot Subject-Driven Generation
• Static Word Embeddings for Sentence Semantic Representation
• VITA: Zero-Shot Value Functions via Test-Time Adaptation of Vision-Language Models
• QGuard:Question-based Zero-shot Guard for Multi-modal LLM Safety
• FeDa4Fair: Client-Level Federated Datasets for Fairness Evaluation
• When Does Multimodality Lead to Better Time Series Forecasting?
• SoMi-ToM: Evaluating Multi-Perspective Theory of Mind in Embodied Social Interactions
• DNN-Based Precoding in RIS-Aided mmWave MIMO Systems With Practical Phase Shift
• Deep Graph Learning for Industrial Carbon Emission Analysis and Policy Impact
• LATTE: Latent Trajectory Embedding for Diffusion-Generated Image Detection
• HumanVideo-MME: Benchmarking MLLMs for Human-Centric Video Understanding
• On the Effectiveness of Methods and Metrics for Explainable AI in Remote Sensing Image Scene Classification
• CADmium: Fine-Tuning Code Language Models for Text-Driven Sequential CAD Design
• Scalable LLM Math Reasoning Acceleration with Low-rank Distillation
• TensorRL-QAS: Reinforcement learning with tensor networks for improved quantum architecture search
• Modeling Saliency Dataset Bias
• DisCO: Reinforcing Large Reasoning Models with Discriminative Constrained Optimization
• scSiameseClu: A Siamese Clustering Framework for Interpreting single-cell RNA Sequencing Data
• Structured Agent Distillation for Large Language Model
• ELEPHANT: Measuring and understanding social sycophancy in LLMs
• Silent Leaks: Implicit Knowledge Extraction Attack on RAG Systems through Benign Queries
• DEBATE, TRAIN, EVOLVE: Self Evolution of Language Model Reasoning
• Causal Interventions Reveal Shared Structure Across English Filler-Gap Constructions
• AudioTrust: Benchmarking the Multifaceted Trustworthiness of Audio Large Language Models
• Find the Fruit: Zero-Shot Sim2Real RL for Occlusion-Aware Plant Manipulation
• LLM Agents for Interactive Exploration of Historical Cadastre Data: Framework and Application to Venice
• Value-Guided Search for Efficient Chain-of-Thought Reasoning
• SelfReflect: Can LLMs Communicate Their Internal Answer Distribution?
• Finite Sample Analysis of Linear Temporal Difference Learning with Arbitrary Features
• ViewSpatial-Bench: Evaluating Multi-perspective Spatial Localization in Vision-Language Models
• Personalized Subgraph Federated Learning with Differentiable Auxiliary Projections
• Linear Attention for Efficient Bidirectional Sequence Modeling
• Adaptive Conformal Guidance for Learning under Uncertainty
• Recent Advances in Large Langauge Model Benchmarks against Data Contamination: From Static to Dynamic Evaluation
• Voting or Consensus? Decision-Making in Multi-Agent Debate
• FANformer: Improving Large Language Models Through Effective Periodicity Modeling
• Answer, Refuse, or Guess? Investigating Risk-Aware Decision Making in Language Models
• Hierarchical Balance Packing: Towards Efficient Supervised Fine-tuning for Long-Context LLM
• Rethinking Diffusion Model in High Dimension
• Revisiting semi-supervised learning in the era of foundation models
• A Survey on SAR ship classification using Deep Learning
• FW-Merging: Scaling Model Merging with Frank-Wolfe Optimization
• Value Profiles for Encoding Human Variation
• CODA: Repurposing Continuous VAEs for Discrete Tokenization
• Enabling Rapid Shared Human-AI Mental Model Alignment via the After-Action Review
• Lobster: A GPU-Accelerated Framework for Neurosymbolic Programming
• Adaptive Rectification Sampling for Test-Time Compute Scaling
• Fair Uncertainty Quantification for Depression Prediction
• Stochastic Layer-wise Learning: Scalable and Efficient Alternative to Backpropagation
• Simple yet Effective Semi-supervised Knowledge Distillation from Vision-Language Models via Dual-Head Optimization
• SSTP: Efficient Sample Selection for Trajectory Prediction
• Composing Global Solutions to Reasoning Tasks via Algebraic Objects in Neural Nets
• FAN: Fourier Analysis Networks
• pEBR: A Probabilistic Approach to Embedding Based Retrieval
• Unlocking Transfer Learning for Open-World Few-Shot Recognition
• BianCang: A Traditional Chinese Medicine Large Language Model
• Learning Semantic Association Rules from Internet of Things Data
• Dynamic Pricing in High-Speed Railways Using Multi-Agent Reinforcement Learning
• Dagger Behind Smile: Fool LLMs with a Happy Ending Story
• A space-decoupling framework for optimization on bounded-rank matrices with orthogonally invariant constraints
• CE-SDWV: Effective and Efficient Concept Erasure for Text-to-Image Diffusion Models via a Semantic-Driven Word Vocabulary
• LFTR: Learning-Free Token Reduction for Multimodal Large Language Models
• Should You Use Your Large Language Model to Explore or Exploit?
• Dual Alignment Maximin Optimization for Offline Model-based RL
• iVISPAR -- An Interactive Visual-Spatial Reasoning Benchmark for VLMs
• Towards Reasoning Ability of Small Language Models
• Solving the Cold Start Problem on One's Own as an End User via Preference Transfer
• A physical approach to qualia and the emergence of conscious observers in qualia space
• Medical Question Summarization with Entity-driven Contrastive Learning
• Tree-Structured Parzen Estimator: Understanding Its Algorithm Components and Their Roles for Better Empirical Performance
• Fast Exact Unlearning for In-Context Learning Data for LLMs
• scCDCG: Efficient Deep Structural Clustering for single-cell RNA-seq via Deep Cut-informed Graph Embedding
• Bird Eye-View to Street-View: A Survey
• Pretrained Hybrids with MAD Skills
• Preemptive Detection and Correction of Misaligned Actions in LLM Agents
• Investigating Long-term Training for Remote Sensing Object Detection
• The Quest for the Right Mediator: Surveying Mechanistic Interpretability Through the Lens of Causal Mediation Analysis
• MENLO: From Preferences to Proficiency - Evaluating and Modeling Native-like Quality Across 47 Languages
• Searching for Difficult-to-Translate Test Examples at Scale
• Learning to See Before Seeing: Demystifying LLM Visual Priors from Language Pre-training
• Learning Generalizable Shape Completion with SIM(3) Equivariance
• OmniRetarget: Interaction-Preserving Data Generation for Humanoid Whole-Body Loco-Manipulation and Scene Interaction
• Stitch: Training-Free Position Control in Multimodal Diffusion Transformers
• Sparse View Tomographic Reconstruction of Elongated Objects using Learned Primal-Dual Networks
• Efficient Dynamic Ensembling for Multiple LLM Experts
• FastCoder: Accelerating Repository-level Code Generation via Efficient Retrieval and Verification
• Memorize or Generalize? Evaluating LLM Code Generation with Code Rewriting
• Teaching AI to Handle Exceptions: Supervised Fine-Tuning with Human-Aligned Judgment
• TAMO: Fine-Grained Root Cause Analysis via Tool-Assisted LLM Agent with Multi-Modality Observation Data in Cloud-Native Systems
• AI Agents vs. Agentic AI: A Conceptual Taxonomy, Applications and Challenges
• Survey: Multi-Armed Bandits Meet Large Language Models
• R1-Code-Interpreter: LLMs Reason with Code via Supervised and Multi-stage Reinforcement Learning
• VS-Bench: Evaluating VLMs for Strategic Abilities in Multi-Agent Environments
• Ascent Fails to Forget
• AdaBlock-dLLM: Semantic-Aware Diffusion LLM Inference via Adaptive Block Size
• ACT: Agentic Classification Tree
• Adaptive Planning for Multi-Attribute Controllable Summarization with Monte Carlo Tree Search
• Attention over Scene Graphs: Indoor Scene Representations Toward CSAI Classification
• On Deepfake Voice Detection - It's All in the Presentation
• Regression Language Models for Code
• VitaBench: Benchmarking LLM Agents with Versatile Interactive Tasks in Real-world Applications
• Indoor/Outdoor Spectrum Sharing Enabled by GNSS-based Classifiers
• The Dragon Hatchling: The Missing Link between the Transformer and Models of the Brain
• MUSE-Explainer: Counterfactual Explanations for Symbolic Music Graph Classification Models
• TAP: Two-Stage Adaptive Personalization of Multi-task and Multi-Modal Foundation Models in Federated Learning
• OceanGym: A Benchmark Environment for Underwater Embodied Agents
• The Unheard Alternative: Contrastive Explanations for Speech-to-Text Models
• Parametric Neural Amp Modeling with Active Learning
• AI-assisted Advanced Propellant Development for Electric Propulsion
• Are Robust LLM Fingerprints Adversarially Robust?
• Deconstructing Self-Bias in LLM-generated Translation Benchmarks
• An Experimental Study on Generating Plausible Textual Explanations for Video Summarization
• 3DiFACE: Synthesizing and Editing Holistic 3D Facial Animation
• Sandbagging in a Simple Survival Bandit Problem
• Finetune Once: Decoupling General & Domain Learning with Dynamic Boosted Annealing
• Point2RBox-v3: Self-Bootstrapping from Point Annotations via Integrated Pseudo-Label Refinement and Utilization
• Representation-Based Data Quality Audits for Audio
• Noise-Guided Transport for Imitation Learning
• QUARTZ : QA-based Unsupervised Abstractive Refinement for Task-oriented Dialogue Summarization
• Feedback Forensics: A Toolkit to Measure AI Personality
• LLM-MCoX: Large Language Model-based Multi-robot Coordinated Exploration and Search
• EditReward: A Human-Aligned Reward Model for Instruction-Guided Image Editing
• SoK: Systematic analysis of adversarial threats against deep learning approaches for autonomous anomaly detection systems in SDN-IoT networks
• TimeScope: Towards Task-Oriented Temporal Grounding In Long Videos
• Vector-Valued Reproducing Kernel Banach Spaces for Neural Networks and Operators
• SDA-PLANNER: State-Dependency Aware Adaptive Planner for Embodied Task Planning
• Efficient and Transferable Agentic Knowledge Graph RAG via Reinforcement Learning
• Game-Time: Evaluating Temporal Dynamics in Spoken Language Models
• SeedPrints: Fingerprints Can Even Tell Which Seed Your Large Language Model Was Trained From
• Real-time Noise Detection and Classification in Single-Channel EEG: A Lightweight Machine Learning Approach for EMG, White Noise, and EOG Artifacts
• On Computing Top-$k$ Simple Shortest Paths from a Single Source
• End-to-End Aspect-Guided Review Summarization at Scale
• Enhancing PINN Performance Through Lie Symmetry Group
• AGOCS -- Accurate Google Cloud Simulator Framework
• Leveraging AI modelling for FDS with Simvue: monitor and optimise for more sustainable simulations
• OWL: Geometry-Aware Spatial Reasoning for Audio Large Language Models
• Bubble, Bubble, AI's Rumble: Why Global Financial Regulatory Incident Reporting is Our Shield Against Systemic Stumbles
• EntroPE: Entropy-Guided Dynamic Patch Encoder for Time Series Forecasting
• Towards Continual Expansion of Data Coverage: Automatic Text-guided Edge-case Synthesis
• Auto-ARGUE: LLM-Based Report Generation Evaluation
• AttriGen: Automated Multi-Attribute Annotation for Blood Cell Datasets
• Optimizing Indoor Environmental Quality in Smart Buildings Using Deep Learning
• Toward an Unbiased Collective Memory for Efficient LLM-Based Agentic 6G Cross-Domain Management
• Comparative Analysis of Ant Colony Optimization and Google OR-Tools for Solving the Open Capacitated Vehicle Routing Problem in Logistics
• Beyond Pixels: Efficient Dataset Distillation via Sparse Gaussian Representation
• Type-Less yet Type-Aware Inductive Link Prediction with Pretrained Language Models
• PerQ: Efficient Evaluation of Multilingual Text Personalization Quality
• User-Centric Communication Service Provision for Edge-Assisted Mobile Augmented Reality
• Accelerating LLM Inference with Precomputed Query Storage
• The Impact of Scaling Training Data on Adversarial Robustness
• From MNIST to ImageNet: Understanding the Scalability Boundaries of Differentiable Logic Gate Networks
• AIM: Adaptive Intervention for Deep Multi-task Learning of Molecular Properties
• Data-Free Continual Learning of Server Models in Model-Heterogeneous Federated learning
• Reconcile Certified Robustness and Accuracy for DNN-based Smoothed Majority Vote Classifier
• R-Log: Incentivizing Log Analysis Capability in LLMs via Reasoning-based Reinforcement Learning
• MHINDR - a DSM5 based mental health diagnosis and recommendation framework using LLM
• VRWKV-Editor: Reducing quadratic complexity in transformer-based video editing
• Learning Egocentric In-Hand Object Segmentation through Weak Supervision from Human Narrations
• PFDepth: Heterogeneous Pinhole-Fisheye Joint Depth Estimation via Distortion-aware Gaussian-Splatted Volumetric Fusion
• Indirect Attention: Turning Context Misalignment into a Feature
• Muon Outperforms Adam in Tail-End Associative Memory Learning
• SeMoBridge: Semantic Modality Bridge for Efficient Few-Shot Adaptation of CLIP
• CEAID: Benchmark of Multilingual Machine-Generated Text Detection Methods for Central European Languages
• Editable Noise Map Inversion: Encoding Target-image into Noise For High-Fidelity Image Manipulation
• Point-It-Out: Benchmarking Embodied Reasoning for Vision Language Models in Multi-Stage Visual Grounding
• Better with Less: Small Proprietary Models Surpass Large Language Models in Financial Transaction Understanding
• CardioForest: An Explainable Ensemble Learning Model for Automatic Wide QRS Complex Tachycardia Diagnosis from ECG
• Learning to Reason as Action Abstractions with Scalable Mid-Training RL
• VELA: An LLM-Hybrid-as-a-Judge Approach for Evaluating Long Image Captions
• Overthinking Reduction with Decoupled Rewards and Curriculum Data Scheduling
• Supporting Creative Ownership through Deep Learning-Based Music Variation
• Distillation of Large Language Models via Concrete Score Matching
• RAE: A Neural Network Dimensionality Reduction Method for Nearest Neighbors Preservation in Vector Search
• S$^2$FS: Spatially-Aware Separability-Driven Feature Selection in Fuzzy Decision Systems
• Training-Free Reward-Guided Image Editing via Trajectory Optimal Control
• More Thought, Less Accuracy? On the Dual Nature of Reasoning in Vision-Language Models
• Knapsack RL: Unlocking Exploration of LLMs via Optimizing Budget Allocation
• Vector sketch animation generation with differentialable motion trajectories
• Efficient On-Policy Reinforcement Learning via Exploration of Sparse Parameter Space
• scUnified: An AI-Ready Standardized Resource for Single-Cell RNA Sequencing Analysis
• RoleConflictBench: A Benchmark of Role Conflict Scenarios for Evaluating LLMs' Contextual Sensitivity
• Capacity-Net-Based RIS Precoding Design without Channel Estimation for mmWave MIMO System
• Deep Reinforcement Learning-Based Precoding for Multi-RIS-Aided Multiuser Downlink Systems with Practical Phase Shift
• EEG-based AI-BCI Wheelchair Advancement: Hybrid Deep Learning with Motor Imagery for Brain Computer Interface
• LD-MoLE: Learnable Dynamic Routing for Mixture of LoRA Experts
• Annotation-Efficient Active Test-Time Adaptation with Conformal Prediction
• HNote: Extending YNote with Hexadecimal Encoding for Fine-Tuning LLMs in Music Modeling
• DeepCodeSeek: Real-Time API Retrieval for Context-Aware Code Generation
• The AI Productivity Index (APEX)
• Towards A Universally Transferable Acceleration Method for Density Functional Theory
• Boundary-to-Region Supervision for Offline Safe Reinforcement Learning
• Controlled Generation for Private Synthetic Text
• Think Less, Label Better: Multi-Stage Domain-Grounded Synthetic Data Generation for Fine-Tuning Large Language Models in Telecommunications
• Dolphin v1.0 Technical Report
• TruthRL: Incentivizing Truthful LLMs via Reinforcement Learning
• Free Lunch Alignment of Text-to-Image Diffusion Models without Preference Image Pairs
• V-HUB: A Visual-Centric Humor Understanding Benchmark for Video LLMs
• Autonomy-Aware Clustering: When Local Decisions Supersede Global Prescriptions
• LLM-RG: Referential Grounding in Outdoor Scenarios using Large Language Models
• MixtureVitae: Open Web-Scale Pretraining Dataset With High Quality Instruction and Reasoning Data Built from Permissive-First Text Sources
• Calibrating Verbalized Confidence with Self-Generated Distractors
• VISOR++: Universal Visual Inputs based Steering for Large Vision Language Models
• Self-Rewarding Rubric-Based Reinforcement Learning for Open-Ended Reasoning
• Steering an Active Learning Workflow Towards Novel Materials Discovery via Queue Prioritization
• Toxicity in Online Platforms and AI Systems: A Survey of Needs, Challenges, Mitigations, and Future Directions
• Vision-Zero: Scalable VLM Self-Improvement via Strategic Gamified Self-Play
• Aligning Multilingual Reasoning with Verifiable Semantics from a High-Resource Expert Model
• Hybrid Approach for Enhancing Lesion Segmentation in Fundus Images
• Probing the Limits of Stylistic Alignment in Vision-Language Models
• AttentionViG: Cross-Attention-Based Dynamic Neighbor Aggregation in Vision GNNs
• K-Prism: A Knowledge-Guided and Prompt Integrated Universal Medical Image Segmentation Model
• Unsupervised Detection of Spatiotemporal Anomalies in PMU Data Using Transformer-Based BiGAN
• Quadratic Programming Approach for Nash Equilibrium Computation in Multiplayer Imperfect-Information Games
• STAC: When Innocent Tools Form Dangerous Chains to Jailbreak LLM Agents
• BaB-prob: Branch and Bound with Preactivation Splitting for Probabilistic Verification of Neural Networks
• YOLO-Based Defect Detection for Metal Sheets
• Data-Efficient Multitask DAgger
• Discontinuous Epitope Fragments as Sufficient Target Templates for Efficient Binder Design
• Translation from Wearable PPG to 12-Lead ECG
• EMO-TTA: Improving Test-Time Adaptation of Audio-Language Models for Speech Emotion Recognition
• Not Wrong, But Untrue: LLM Overconfidence in Document-Based Queries
• DeepFake Detection in Dyadic Video Calls using Point of Gaze Tracking
• XR Blocks: Accelerating Human-centered AI + XR Innovation
• Economic Competition, EU Regulation, and Executive Orders: A Framework for Discussing AI Policy Implications in CS Courses
• FlashOmni: A Unified Sparse Attention Engine for Diffusion Transformers
• From Faithfulness to Correctness: Generative Reward Models that Think Critically
• Rethinking Parameter Sharing for LLM Fine-Tuning with Multiple LoRAs
• Emotion-Aligned Generation in Diffusion Text to Speech Models via Preference-Guided Optimization
• Polychromic Objectives for Reinforcement Learning
• Beyond Noisy-TVs: Noise-Robust Exploration Via Learning Progress Monitoring
• Joint Embeddings Go Temporal
• Multi-patch isogeometric neural solver for partial differential equations on computer-aided design domains
• PIPer: On-Device Environment Setup via Online Reinforcement Learning
• Effectiveness of Large Language Models in Simulating Regional Psychological Structures: An Empirical Examination of Personality and Subjective Well-being
• Artificial Authority: From Machine Minds to Political Alignments. An Experimental Analysis of Democratic and Autocratic Biases in Large-Language Models
• ClustRecNet: A Novel End-to-End Deep Learning Framework for Clustering Algorithm Recommendation
• A Measurement Study of Model Context Protocol
• AI in Pakistani Schools: Adoption, Usage, and Perceived Impact among Educators
• Learning Relationships Between Separate Audio Tracks for Creative Applications
• Automatically Generating Web Applications from Requirements Via Multi-Agent Test-Driven Development
• Scaling Behaviors of LLM Reinforcement Learning Post-Training: An Empirical Study in Mathematical Reasoning
• Uncertainty-Aware Generative Oversampling Using an Entropy-Guided Conditional Variational Autoencoder
• VisualOverload: Probing Visual Understanding of VLMs in Really Dense Scenes
• From Internal Representations to Text Quality: A Geometric Approach to LLM Evaluation
• Generative Value Conflicts Reveal LLM Priorities
• Cold-Start Active Correlation Clustering
• Let Physics Guide Your Protein Flows: Topology-aware Unfolding and Generation
• Predicting Training Re-evaluation Curves Enables Effective Data Curriculums for LLMs
• SpinBench: Perspective and Rotation as a Lens on Spatial Reasoning in VLMs
• A Deep Learning Approach for Spatio-Temporal Forecasting of InSAR Ground Deformation in Eastern Ireland
• A Cartography of Open Collaboration in Open Source AI: Mapping Practices, Motivations, and Governance in 14 Open Large Language Model Projects
• Reinforcement Learning-Guided Chain-of-Draft for Token-Efficient Code Generation
• Comprehensive Analysis of VQC for Financial Fraud Detection: A Comparative Study of Quantum Encoding Techniques and Architectural Optimizations
• Protocode: Prototype-Driven Interpretability for Code Generation in LLMs
• BuildBench: Benchmarking LLM Agents on Compiling Real-World Open-Source Software
• BEV-VLM: Trajectory Planning via Unified BEV Abstraction
• Knowledge distillation through geometry-aware representational alignment
• The Sandbox Configurator: A Framework to Support Technical Assessment in AI Regulatory Sandboxes
• Artificial Intelligence-Powered Assessment Framework for Skill-Oriented Engineering Lab Education
• How Effective Are Time-Series Models for Rainfall Nowcasting? A Comprehensive Benchmark for Rainfall Nowcasting Incorporating PWV Data
• From NL2SQL to NL2GeoSQL: GeoSQL-Eval for automated evaluation of LLMs on PostGIS queries
• Cognifying Education: Mapping AI's transformative role in emotional, creative, and collaborative learning
• Dynamic Policy Induction for Adaptive Prompt Optimization: Bridging the Efficiency-Accuracy Gap via Lightweight Reinforcement Learning
• A Weather Foundation Model for the Power Grid
• InfMasking: Unleashing Synergistic Information by Contrastive Multimodal Interactions
• DNABERT-2: Fine-Tuning a Genomic Language Model for Colorectal Gene Enhancer Classification
• VoiceBridge: Designing Latent Bridge Models for General Speech Restoration at Scale
• Devstral: Fine-tuning Language Models for Coding Agent Applications
• APRIL: API Synthesis with Automatic Prompt Optimization and Reinforcement Learning
• Towards Repository-Level Program Verification with Large Language Models
• Generating High-Quality Datasets for Code Editing via Open-Source Language Models
• Spectral Logit Sculpting: Adaptive Low-Rank Logit Transformation for Controlled Text Generation
• Multi-level Diagnosis and Evaluation for Robust Tabular Feature Engineering with Large Language Models
• STCast: Adaptive Boundary Alignment for Global and Regional Weather Forecasting
• Six Sigma For Neural Networks: Taguchi-based optimization
• On-the-Fly Adaptation to Quantization: Configuration-Aware LoRA for Efficient Fine-Tuning of Quantized LLMs
• Learning to Condition: A Neural Heuristic for Scalable MPE Inference
• Enhancing Linear Attention with Residual Learning
• Energy Guided Geometric Flow Matching
• FedCLF - Towards Efficient Participant Selection for Federated Learning in Heterogeneous IoV Networks
• Machine Learning for Pattern Detection in Printhead Nozzle Logging
• Quantum est in Libris: Navigating Archives with GenAI, Uncovering Tension Between Preservation and Innovation
• PALADIN: Self-Correcting Language Model Agents to Cure Tool-Failure Cases
• HAMMER: Hamiltonian Curiosity Augmented Large Language Model Reinforcement
• A Benchmark for Localizing Code and Non-Code Issues in Software Projects
• Zero-Shot Decentralized Federated Learning
• Extreme Self-Preference in Language Models
• STaR-Attack: A Spatio-Temporal and Narrative Reasoning Attack Framework for Unified Multimodal Understanding and Generation Models
• The Average Patient Fallacy
• TVS Sidekick: Challenges and Practical Insights from Deploying Large Language Models in the Enterprise
• Combining Knowledge Graphs and NLP to Analyze Instant Messaging Data in Criminal Investigations
• OffTopicEval: When Large Language Models Enter the Wrong Chat, Almost Always!
• SCUBA: Salesforce Computer Use Benchmark
• Rearchitecting Datacenter Lifecycle for AI: A TCO-Driven Framework
• HilbertA: Hilbert Attention for Image Generation with Diffusion Models
• Probing the Critical Point (CritPt) of AI Reasoning: a Frontier Physics Research Benchmark
• Fairness Testing in Retrieval-Augmented Generation: How Small Perturbations Reveal Bias in Small Language Models
• Fine-tuning Behavioral Cloning Policies with Preference-Based Reinforcement Learning
• TimeRewarder: Learning Dense Reward from Passive Videos via Frame-wise Temporal Distance
• Branching Out: Broadening AI Measurement and Evaluation with Measurement Trees
• Diagnosing and Addressing Pitfalls in KG-RAG Datasets: Toward More Reliable Benchmarking
• FMIP: Joint Continuous-Integer Flow For Mixed-Integer Linear Programming
• AdaptCache: KV Cache Native Storage Hierarchy for Low-Delay and High-Quality Language Model Serving
• Beyond the Algorithm: A Field Guide to Deploying AI Agents in Clinical Practice
• 90% Faster, 100% Code-Free: MLLM-Driven Zero-Code 3D Game Development
• 'Too much alignment; not enough culture': Re-balancing cultural alignment practices in LLMs
• LLM Agents for Knowledge Discovery in Atomic Layer Processing
• Human-Centered Evaluation of RAG outputs: a framework and questionnaire for human-AI collaboration
• Diversity-Incentivized Exploration for Versatile Reasoning
• Benchmarking Deep Learning Convolutions on Energy-constrained CPUs
• SlimPack: Fine-Grained Asymmetric Packing for Balanced and Efficient Variable-Length LLM Training
• ExoPredicator: Learning Abstract Models of Dynamic Worlds for Robot Planning
• Interactive Learning for LLM Reasoning
• AI Playing Business Games: Benchmarking Large Language Models on Managerial Decision-Making in Dynamic Simulations
• SafeBehavior: Simulating Human-Like Multistage Reasoning to Mitigate Jailbreak Attacks in Large Language Models
• How Far Do Time Series Foundation Models Paint the Landscape of Real-World Benchmarks ?
• Your Agent May Misevolve: Emergent Risks in Self-evolving LLM Agents
• MC-GNNAS-Dock: Multi-criteria GNN-based Algorithm Selection for Molecular Docking
• Commmunication-Efficient and Accurate Approach for Aggregation in Federated Low-Rank Adaptation
• OntoAligner Meets Knowledge Graph Embedding Aligners
• Transformer Classification of Breast Lesions: The BreastDCEDL_AMBL Benchmark Dataset and 0.92 AUC Baseline
• CIMNAS: A Joint Framework for Compute-In-Memory-Aware Neural Architecture Search
• Lita: Light Agent Uncovers the Agentic Coding Capabilities of LLMs
• SafeMind: Benchmarking and Mitigating Safety Risks in Embodied LLM Agents
• DeepJSONEval: Benchmarking Complex Nested JSON Data Mining for Large Language Models
• KIRETT: Smart Integration of Vital Signs Data for Intelligent Decision Support in Rescue Scenarios
• Quantitative Evaluation of KIRETT Wearable Demonstrator for Rescue Operations
• Boosting Process-Correct CoT Reasoning by Modeling Solvability of Multiple-Choice QA
• NuRisk: A Visual Question Answering Dataset for Agent-Level Risk Assessment in Autonomous Driving
• Automated Model Discovery via Multi-modal & Multi-step Pipeline
• RoRecomp: Enhancing Reasoning Efficiency via Rollout Response Recomposition in Reinforcement Learning
• Scalable and Robust LLM Unlearning by Correcting Responses with Retrieved Exclusions
• Towards Unified Multimodal Misinformation Detection in Social Media: A Benchmark Dataset and Baseline
• Towards Human Engagement with Realistic AI Combat Pilots
• CoLLM-NAS: Collaborative Large Language Models for Efficient Knowledge-Guided Neural Architecture Search
• Evaluating the Use of Large Language Models as Synthetic Social Agents in Social Science Research
• SafeEvalAgent: Toward Agentic and Self-Evolving Safety Evaluation of LLMs
• MEDAKA: Construction of Biomedical Knowledge Graphs Using Large Language Models
• LMILAtt: A Deep Learning Model for Depression Detection from Social Media Users Enhanced by Multi-Instance Learning Based on Attention Mechanism
• On Explaining Proxy Discrimination and Unfairness in Individual Decisions Made by AI Systems
• GroundSight: Augmenting Vision-Language Models with Grounding Information and De-hallucination
• SING-SQL: A Synthetic Data Generation Framework for In-Domain Text-to-SQL Translation
• Collaborative Compression for Large-Scale MoE Deployment on Edge
• ScheduleMe: Multi-Agent Calendar Assistant
• Cooperative Autonomous Driving in Diverse Behavioral Traffic: A Heterogeneous Graph Reinforcement Learning Approach
• NePTune: A Neuro-Pythonic Framework for Tunable Compositional Reasoning on Vision-Language
• Thinking Sparks!: Emergent Attention Heads in Reasoning Models During Post Training
• Galton's Law of Mediocrity: Why Large Language Models Regress to the Mean and Fail at Creativity in Advertising
• Planner-R1: Reward Shaping Enables Efficient Agentic RL with Smaller LLMs
• Deontic Argumentation
• PUREVQ-GAN: Defending Data Poisoning Attacks through Vector-Quantized Bottlenecks
• Chain-in-Tree: Back to Sequential Reasoning in LLM Tree Search
• HiStyle: Hierarchical Style Embedding Predictor for Text-Prompt-Guided Controllable Speech Synthesis
• ASGuard: Activation-Scaling Guard to Mitigate Targeted Jailbreaking Attack
• Aging Decline in Basketball Career Trend Prediction Based on Machine Learning and LSTM Model
• RadOnc-GPT: An Autonomous LLM Agent for Real-Time Patient Outcomes Labeling at Scale
• Learning to Interact in World Latent for Team Coordination
• Evaluating Foundation Models with Pathological Concept Learning for Kidney Cancer
• A(I)nimism: Re-enchanting the World Through AI-Mediated Object Interaction
• Radiology's Last Exam (RadLE): Benchmarking Frontier Multimodal AI Against Human Experts and a Taxonomy of Visual Reasoning Errors in Radiology
• IRIS: Intrinsic Reward Image Synthesis
• Skip-It? Theoretical Conditions for Layer Skipping in Vision-Language Models
• ATLAS: Constraints-Aware Multi-Agent Collaboration for Real-World Travel Planning
• Building the EHR Foundation Model via Next Event Prediction
• Causal Autoencoder-like Generation of Feedback Fuzzy Cognitive Maps with an LLM Agent
• Hybrid Reward Normalization for Process-supervised Non-verifiable Agentic Tasks
• Echoes of Humanity: Exploring the Perceived Humanness of AI Music
• A Framework for Studying AI Agent Behavior: Evidence from Consumer Choice Experiments
• SMS: Self-supervised Model Seeding for Verification of Machine Unlearning
• SOCK: A Benchmark for Measuring Self-Replication in Large Language Models
• AutoLabs: Cognitive Multi-Agent Systems with Self-Correction for Autonomous Chemical Experimentation
• Iterative Residual Cross-Attention Mechanism: An Integrated Approach for Audio-Visual Navigation Tasks
• Landmark-Guided Knowledge for Vision-and-Language Navigation
• Dive into the Agent Matrix: A Realistic Evaluation of Self-Replication Risk in LLM Agents
• Spontaneous High-Order Generalization in Neural Theory-of-Mind Networks
• SynthPert: Enhancing LLM Biological Reasoning via Synthetic Reasoning Traces for Cellular Perturbation Prediction
• Structural Reward Model: Enhancing Interpretability, Efficiency, and Scalability in Reward Modeling
• Where LLM Agents Fail and How They can Learn From Failures
• From Perception to Cognition: A Survey of Vision-Language Interactive Reasoning in Multimodal Large Language Models
• Saliency Guided Longitudinal Medical Visual Question Answering
• Boolean Satisfiability via Imitation Learning
• Adaptive Test-Time Reasoning via Reward-Guided Dual-Phase Search
• RADAR: Reasoning-Ability and Difficulty-Aware Routing for Reasoning LLMs
• The Open Syndrome Definition
• GESA: Graph-Enhanced Semantic Allocation for Generalized, Fair, and Explainable Candidate-Role Matching
• DeepSearch: Overcome the Bottleneck of Reinforcement Learning with Verifiable Rewards via Monte Carlo Tree Search
• Plug-and-Play Emotion Graphs for Compositional Prompting in Zero-Shot Speech Emotion Recognition
• TDHook: A Lightweight Framework for Interpretability
• Message passing-based inference in an autoregressive active inference agent
• Understanding Generative Recommendation with Semantic IDs from a Model-scaling View
• Beyond Static Retrieval: Opportunities and Pitfalls of Iterative Retrieval in GraphRAG
• Blueprint-Bench: Comparing spatial intelligence of LLMs, agents and image models
• The Causal Abstraction Network: Theory and Learning
• A Formal Comparison Between Chain-of-Thought and Latent Thought
• Neo-Grounded Theory: A Methodological Innovation Integrating High-Dimensional Vector Clustering and Multi-Agent Collaboration for Qualitative Research
• Memory Management and Contextual Consistency for Long-Running Low-Code Agents
• Fact Grounded Attention: Eliminating Hallucination in Large Language Models Through Attention Level Knowledge Integration
• Language Model Planning from an Information Theoretic Perspective
• RADAR: A Risk-Aware Dynamic Multi-Agent Framework for LLM Safety Evaluation via Role-Specialized Collaboration
• RL in the Wild: Characterizing RLVR Training in LLM Deployment
• Toward Causal-Visual Programming: Enhancing Agentic Reasoning in Low-Code Environments
• ID-RAG: Identity Retrieval-Augmented Generation for Long-Horizon Persona Coherence in Generative Agents
• Flash-Searcher: Fast and Effective Web Agents via DAG-Based Parallel Execution