AI Research News Feeds for October 17th, 2025

AI RESEARCH PAPERS & ACADEMIC SOURCES

• ScholarBench: A Bilingual Benchmark for Abstraction, Comprehension, and Reasoning Evaluation in Academic Contexts
• Multilinguality Does not Make Sense: Investigating Factors Behind Zero-Shot Transfer in Sense-Aware Tasks
• Are LLMs Stable Formal Logic Translators in Logical Reasoning Across Linguistically Diversified Texts?
• Thunder-DeID: Accurate and Efficient De-identification Framework for Korean Court Judgments
• EasyNER: A Customizable Easy-to-Use Pipeline for Deep Learning- and Dictionary-based Named Entity Recognition from Medical and Life Science Text
• CAP: Evaluation of Persuasive and Creative Image Generation
• Seeing Through Green: Text-Based Classification and the Firm's Returns from Green Patents
• MultiFoodhat: A potential new paradigm for intelligent food quality inspection
• NAPPure: Adversarial Purification for Robust Image Classification under Non-Additive Perturbations
• Vgent: Graph-based Retrieval-Reasoning-Augmented Generation For Long Video Understanding
• Synchronization of Multiple Videos
• Capture, Canonicalize, Splat: Zero-Shot 3D Gaussian Avatars from Unstructured Phone Images
• cubic: CUDA-accelerated 3D Bioimage Computing
• LOTA: Bit-Planes Guided AI-Generated Image Detection
• PIA: Deepfake Detection Using Phoneme-Temporal and Identity-Dynamic Analysis
• Event Interval Modulation: A Novel Scheme for Event-based Optical Camera Communication
• MACE: Mixture-of-Experts Accelerated Coordinate Encoding for Large-Scale Scene Localization and Rendering
• Identity-Preserving Image-to-Video Generation via Reward-Guided Optimization
• Identity-GRPO: Optimizing Multi-Human Identity-preserving Video Generation via Reinforcement Learning
• MatchAttention: Matching the Relative Positions for High-Resolution Cross-View Matching
• Experimental Demonstration of Event-based Optical Camera Communication in Long-Range Outdoor Environment
• GauSSmart: Enhanced 3D Reconstruction through 2D Foundation Models and Geometric Filtering
• CLEAR: Causal Learning Framework For Robust Histopathology Tumor Detection Under Out-Of-Distribution Shifts
• A Multi-domain Image Translative Diffusion StyleGAN for Iris Presentation Attack Detection
• Vision-Centric Activation and Coordination for Multimodal Large Language Models
• Leveraging Cycle-Consistent Anchor Points for Self-Supervised RGB-D Registration
• Spatial Preference Rewarding for MLLMs Spatial Understanding
• DOS: Directional Object Separation in Text Embeddings for Multi-Object Image Generation
• DRBD-Mamba for Robust and Efficient Brain Tumor Segmentation with Analytical Insights
• DCMIL: A Progressive Representation Learning Model of Whole Slide Images for Cancer Prognosis Analysis
• Real-Time Neural Video Compression with Unified Intra and Inter Coding
• Structured Universal Adversarial Attacks on Object Detection for Video Sequences
• Unsupervised Deep Generative Models for Anomaly Detection in Neuroimaging: A Systematic Scoping Review
• Pruning Overparameterized Multi-Task Networks for Degraded Web Image Restoration
• Grazing Detection using Deep Learning and Sentinel-2 Time Series Data
• Vision Mamba for Permeability Prediction of Porous Media
• PaddleOCR-VL: Boosting Multilingual Document Parsing via a 0.9B Ultra-Compact Vision-Language Model
• Towards Generalist Intelligence in Dentistry: Vision Foundation Models for Oral and Maxillofacial Radiology
• Acquisition of interpretable domain information during brain MR image harmonization for content-based image retrieval
• Exploring Image Representation with Decoupled Classical Visual Descriptors
• Exploring Cross-Modal Flows for Few-Shot Learning
• Consistent text-to-image generation via scene de-contextualization
• Eyes Wide Open: Ego Proactive Video-LLM for Streaming Video
• BalanceGS: Algorithm-System Co-design for Efficient 3D Gaussian Splatting Training on GPU
• CALM-Net: Curvature-Aware LiDAR Point Cloud-based Multi-Branch Neural Network for Vehicle Re-Identification
• Hierarchical Re-Classification: Combining Animal Classification Models with Vision Transformers
• Zero-Shot Wildlife Sorting Using Vision Transformers: Evaluating Clustering and Continuous Similarity Ordering
• Shot2Tactic-Caption: Multi-Scale Captioning of Badminton Videos for Tactical Understanding
• Efficient Video Sampling: Pruning Temporally Redundant Tokens for Faster VLM Inference
• Adapting Self-Supervised Representations as a Latent Space for Efficient Generation
• SteeringTTA: Guiding Diffusion Trajectories for Robust Test-Time-Adaptation
• EuroMineNet: A Multitemporal Sentinel-2 Benchmark for Spatiotemporal Mining Footprint Analysis in the European Union (2015-2024)
• WeCKD: Weakly-supervised Chained Distillation Network for Efficient Multimodal Medical Imaging
• VTimeCoT: Thinking by Drawing for Video Temporal Grounding and Reasoning
• Leveraging Learned Image Prior for 3D Gaussian Compression
• Cross-Layer Feature Self-Attention Module for Multi-Scale Object Detection
• Free-Grained Hierarchical Recognition
• LightQANet: Quantized and Adaptive Feature Learning for Low-Light Image Enhancement
• MoCom: Motion-based Inter-MAV Visual Communication Using Event Vision and Spiking Neural Networks
• CoT-PL: Visual Chain-of-Thought Reasoning Meets Pseudo-Labeling for Open-Vocabulary Object Detection
• FraQAT: Quantization Aware Training with Fractional bits
• Scaling Tumor Segmentation: Best Lessons from Real and Synthetic Data
• QDepth-VLA: Quantized Depth Prediction as Auxiliary Supervision for Vision-Language-Action Models
• ImagerySearch: Adaptive Test-Time Search for Video Generation Beyond Semantic Dependency Constraints
• Multi-modal video data-pipelines for machine learning with minimal human supervision
• TOUCH: Text-guided Controllable Generation of Free-Form Hand-Object Interactions
• BADAS: Context Aware Collision Prediction Using Real-World Dashcam Data
• ScaleWeaver: Weaving Efficient Controllable T2I Generation with Multi-Scale Reference Attention
• Leveraging Multimodal LLM Descriptions of Activity for Explainable Semi-Supervised Video Anomaly Detection
• 3D Scene Prompting for Scene-Consistent Camera-Controllable Video Generation
• OmniMotion: Multimodal Motion Generation with Continuous Masked Autoregression
• RainDiff: End-to-end Precipitation Nowcasting Via Token-wise Attention Diffusion
• ChangingGrounding: 3D Visual Grounding in Changing Scenes
• Ponimator: Unfolding Interactive Pose for Versatile Human-human Interaction Animation
• Deep Compositional Phase Diffusion for Long Motion Sequence Generation
• GOPLA: Generalizable Object Placement Learning via Synthetic Augmentation of Human Arrangement
• From Language to Locomotion: Retargeting-free Humanoid Control via Motion Latent Guidance
• Neural Plasticity-Inspired Multimodal Foundation Model for Earth Observation
• Incomplete Multimodal Industrial Anomaly Detection via Cross-Modal Distillation
• AttenCraft: Attention-guided Disentanglement of Multiple Concepts for Text-to-Image Customization
• Multi-level Reliable Guidance for Unpaired Multi-view Clustering
• Shape of Motion: 4D Reconstruction from a Single Video
• GRAB: A Challenging GRaph Analysis Benchmark for Large Multimodal Models
• Impact of Regularization on Calibration and Robustness: from the Representation Space Perspective
• Perspective-Aware Teaching: Adapting Knowledge for Heterogeneous Distillation
• HuGDiffusion: Generalizable Single-Image Human Rendering via 3D Gaussian Diffusion
• LinPrim: Linear Primitives for Differentiable Volumetric Rendering
• TRACE: Your Diffusion Model is Secretly an Instance Edge Detector
• Falcon: A Remote Sensing Vision-Language Foundation Model (Technical Report)
• Spot the Fake: Large Multimodal Model-Based Synthetic Image Detection with Artifact Explanation
• OmnimatteZero: Fast Training-free Omnimatte with Pre-trained Video Diffusion Models
• Training-Free Personalization via Retrieval and Reasoning on Fingerprints
• 3DOT: Texture Transfer for 3DGS Objects from a Single Reference Image
• On Large Multimodal Models as Open-World Image Classifiers
• KeyVID: Keyframe-Aware Video Diffusion for Audio-Synchronized Visual Animation
• ForensicHub: A Unified Benchmark & Codebase for All-Domain Fake Image Detection and Localization
• TinyRS-R1: Compact Multimodal Language Model for Remote Sensing
• SphereDrag: Spherical Geometry-Aware Panoramic Image Editing
• MetaQAP - A Meta-Learning Approach for Quality-Aware Pretraining in Image Quality Assessment
• CL-Splats: Continual Learning of Gaussian Splatting with Local Optimization
• ComposeMe: Attribute-Specific Image Prompts for Controllable Human Image Generation
• SVAG-Bench: A Large-Scale Benchmark for Multi-Instance Spatio-temporal Video Action Grounding
• CymbaDiff: Structured Spatial Diffusion for Sketch-based 3D Semantic Urban Scene Generation
• Group-Wise Optimization for Self-Extensible Codebooks in Vector Quantized Models
• High Semantic Features for the Continual Learning of Complex Emotions: a Lightweight Solution
• OmniGaze: Reward-inspired Generalizable Gaze Estimation In The Wild
• Reasoning in Space via Grounding in the World
• TABSurfer: a Hybrid Deep Learning Architecture for Subcortical Segmentation
• Pruning Sparse Tensor Neural Networks Enables Deep Learning for 3D Ultrasound Localization Microscopy
• Deep Few-view High-resolution Photon-counting CT at Halved Dose for Extremity Imaging
• From Explainability to Action: A Generative Operational Framework for Integrating XAI in Clinical Mental Health Screening
• Quechua Speech Datasets in Common Voice: The Case of Puno Quechua
• TextBandit: Evaluating Probabilistic Reasoning in LLMs Through Language-Only Decision Tasks
• Too Open for Opinion? Embracing Open-Endedness in Large Language Models for Social Simulation
• The Harder The Better: Maintaining Supervised Fine-tuning Generalization with Less but Harder Data
• Attribution Quality in AI-Generated Content:Benchmarking Style Embeddings and LLM Judges
• RAID: Refusal-Aware and Integrated Decoding for Jailbreaking LLMs
• Investigating Political and Demographic Associations in Large Language Models Through Moral Foundations Theory
• LLM Prompt Duel Optimizer: Efficient Label-Free Prompt Optimization
• Interpreting the Latent Structure of Operator Precedence in Language Models
• RAGCap-Bench: Benchmarking Capabilities of LLMs in Agentic Retrieval Augmented Generation Systems
• Element2Vec: Build Chemical Element Representation from Text for Property Prediction
• Optimal Aggregation of LLM and PRM Signals for Efficient Test-Time Scaling
• FACTS: Table Summarization via Offline Template Generation with Agentic Workflows
• An LLM-Powered AI Agent Framework for Holistic IoT Traffic Interpretation
• BioMedSearch: A Multi-Source Biomedical Retrieval Framework Based on LLMs
• Robust or Suggestible? Exploring Non-Clinical Induction in LLM Drug-Safety Decisions
• FinDeepResearch: Evaluating Deep Research Agents in Rigorous Financial Analysis
• Readers Prefer Outputs of AI Trained on Copyrighted Books over Expert Human Writers
• The German Commons - 154 Billion Tokens of Openly Licensed Text for German Language Models
• CRaFT: An Explanation-Based Framework for Evaluating Cultural Reasoning in Multilingual Language Models
• Quantifying Phonosemantic Iconicity Distributionally in 6 Languages
• ERGO: Entropy-guided Resetting for Generation Optimization in Multi-turn Language Models
• DROID: Dual Representation for Out-of-Scope Intent Detection
• Building a Macedonian Recipe Dataset: Collection, Parsing, and Comparative Analysis
• RLSR: Reinforcement Learning with Supervised Reward Outperforms SFT in Instruction Following
• MoM: Mixtures of Scenario-Aware Document Memories for Retrieval-Augmented Generation Systems
• Rewriting History: A Recipe for Interventional Analyses to Study Data Effects on Model Behavior
• Retrofitting Small Multilingual Models for Retrieval: Matching 7B Performance with 300M Parameters
• Qwen3Guard Technical Report
• Rethinking Schema Linking: A Context-Aware Bidirectional Retrieval Approach for Text-to-SQL
• MathMist: A Parallel Multilingual Benchmark Dataset for Mathematical Problem Solving and Reasoning
• On the Ability of LLMs to Handle Character-Level Perturbations: How Well and How?
• Suicidal Comment Tree Dataset: Enhancing Risk Assessment and Prediction Through Contextual Analysis
• Your Next Token Prediction: A Multilingual Benchmark for Personalized Response Generation
• Explore to Evolve: Scaling Evolved Aggregation Logic via Proactive Online Exploration for Deep Research Agents
• Natural Language Tools: A Natural Language Approach to Tool Calling In Large Language Agents
• Efficient Seq2seq Coreference Resolution Using Entity Representations
• Assessing Socio-Cultural Alignment and Technical Safety of Sovereign LLMs
• Intent Clustering with Shared Pseudo-Labels
• Semantic Prosody in Machine Translation: the English-Chinese Case of Passive Structures
• Speculative Model Risk in Healthcare AI: Using Storytelling to Surface Unintended Harms
• AutoRubric-R1V: Rubric-Based Generative Rewards for Faithful Multimodal Reasoning
• Pluto: A Benchmark for Evaluating Efficiency of LLM-generated Hardware Code
• Supervised Fine-Tuning or Contrastive Learning? Towards Better Multimodal LLM Reranking
• Rewiring Experts on the Fly:Continuous Rerouting for Better Online Adaptation in Mixture-of-Expert models
• Midtraining Bridges Pretraining and Posttraining Distributions
• Harmonizing Diverse Models: A Layer-wise Merging Strategy for Consistent Generation
• AI-Powered Early Diagnosis of Mental Health Disorders from Real-World Clinical Conversations
• Joint Modeling of Big Five and HEXACO for Multimodal Apparent Personality-trait Recognition
• Talking Points: Describing and Localizing Pixels
• You May Speak Freely: Improving the Fine-Grained Visual Recognition Capabilities of Multimodal Large Language Models with Answer Extraction
• MathCanvas: Intrinsic Visual Chain-of-Thought for Multimodal Mathematical Reasoning
• AAVENUE: Detecting LLM Biases on NLU Tasks in AAVE via a Novel Benchmark
• Multi-Perspective Stance Detection
• Why We Build Local Large Language Models: An Observational Analysis from 35 Japanese and Multilingual LLMs
• Women, Infamous, and Exotic Beings: A Comparative Study of Honorific Usages in Wikipedia and LLMs for Bengali and Hindi
• Following the Autoregressive Nature of LLM Embeddings via Compression and Alignment
• Sentence Smith: Controllable Edits for Evaluating Text Embeddings
• Revisiting Prompt Optimization with Large Reasoning Models-A Case Study on Event Extraction
• Can Pre-training Indicators Reliably Predict Fine-tuning Outcomes of LLMs?
• Offline Reinforcement Learning via Inverse Optimization
• Strategyproof Reinforcement Learning from Human Feedback
• Just One Layer Norm Guarantees Stable Extrapolation
• Electrostatics from Laplacian Eigenbasis for Neural Network Interatomic Potentials
• Uni-LoRA: One Vector is All You Need
• TimeRecipe: A Time-Series Forecasting Recipe via Benchmarking Module Level Effectiveness
• Calibrated Predictive Lower Bounds on Time-to-Unsafe-Sampling in LLMs
• Byzantine Failures Harm the Generalization of Robust Distributed Learning Algorithms More Than Data Poisoning
• BoltzNCE: Learning Likelihoods for Boltzmann Generation with Stochastic Interpolants and Noise Contrastive Estimation
• Flows and Diffusions on the Neural Manifold
• The Principle of Uncertain Maximum Entropy
• SoK: Adversarial Evasion Attacks Practicality in NIDS Domain and the Impact of Dynamic Learning
• An analysis of the derivative-free loss method for solving PDEs
• Integrating feature selection and regression methods with technical indicators for predicting Apple Inc. stock prices
• Improving Intrusion Detection with Domain-Invariant Representation Learning in Latent Space
• SOHES: Self-supervised Open-world Hierarchical Entity Segmentation
• Approximation Rates for Shallow ReLU$^k$ Neural Networks on Sobolev Spaces via the Radon Transform
• Generating High Dimensional User-Specific Wireless Channels using Diffusion Models
• Uncertainty Quantification with the Empirical Neural Tangent Kernel
• Minimax Optimal Kernel Two-Sample Tests with Random Features
• Technical and legal aspects of federated learning in bioinformatics: applications, challenges and opportunities
• ELASTIC: Efficient Once For All Iterative Search for Object Detection on Microcontrollers
• SPIRIT: Patching Speech Language Models against Jailbreak Attacks
• Adaptive Set-Mass Calibration with Conformal Prediction
• Mitigating Hallucinations in Vision-Language Models through Image-Guided Head Suppression
• Adversarial Disentanglement by Backpropagation with Physics-Informed Variational Autoencoder
• Mapping Farmed Landscapes from Remote Sensing
• Online Continual Learning via Spiking Neural Networks with Sleep Enhanced Latent Replay
• Real-Time Adaptive Motion Planning via Point Cloud-Guided, Energy-Based Diffusion and Potential Fields
• Protenix-Mini+: efficient structure prediction model with scalable pairformer
• SimULi: Real-Time LiDAR and Camera Simulation with Unscented Transforms
• MimicKit: A Reinforcement Learning Framework for Motion Imitation and Control
• Attention-Aided MMSE for OFDM Channel Estimation: Learning Linear Filters with Attention
• iQUEST: An Iterative Question-Guided Framework for Knowledge Base Question Answering
• When Style Breaks Safety: Defending LLMs Against Superficial Style Alignment
• KScope: A Framework for Characterizing the Knowledge Status of Language Models
• TAI3: Testing Agent Integrity in Interpreting User Intent
• SoK: Evaluating Jailbreak Guardrails for Large Language Models
• Subspace-Boosted Model Merging
• LLM-guided Chemical Process Optimization with a Multi-Agent Approach
• VALID-Mol: a Systematic Framework for Validated LLM-Assisted Molecular Design
• A Clinically-Grounded Two-Stage Framework for Renal CT Report Generation
• TopoStreamer: Temporal Lane Segment Topology Reasoning in Autonomous Driving
• Gemini 2.5: Pushing the Frontier with Advanced Reasoning, Multimodality, Long Context, and Next Generation Agentic Capabilities
• Prompt Perturbations Reveal Human-Like Biases in Large Language Model Survey Responses
• Why is Your Language Model a Poor Implicit Reward Model?
• HANS-Net: Hyperbolic Convolution and Adaptive Temporal Attention for Accurate and Generalizable Liver and Tumor Segmentation in CT Imaging
• Causal Language Control in Multilingual Transformers via Sparse Feature Steering
• Rethinking Purity and Diversity in Multi-Behavior Sequential Recommendation from the Frequency Perspective
• Merge-of-Thought Distillation
• Chiplet-Based RISC-V SoC with Modular AI Acceleration
• Large Language Models for Real-World IoT Device Identification
• Multi-View Semi-Supervised Label Distribution Learning with Local Structure Complementarity
• Weight Weaving: Parameter Pooling for Data-Free Model Merging
• LTR-ICD: A Learning-to-Rank Approach for Automatic ICD Coding
• Distributional Consistency Loss: Beyond Pointwise Data Terms in Inverse Problems
• BitNet Distillation
• Noise-Adaptive Layerwise Learning Rates: Accelerating Geometry-Aware Optimization for Deep Neural Network Training
• CausalVerse: Benchmarking Causal Representation Learning with Configurable High-Fidelity Simulations
• FedHFT: Efficient Federated Finetuning with Heterogeneous Edge Clients
• Neural Network approximation power on homogeneous and heterogeneous reaction-diffusion equations
• TENDE: Transfer Entropy Neural Diffusion Estimation
• Near-Optimal Regret-Queue Length Tradeoff in Online Learning for Two-Sided Markets
• Briding Diffusion Posterior Sampling and Monte Carlo methods: a survey
• Neural Network-enabled Domain-consistent Robust Optimisation for Global CO$_2$ Reduction Potential of Gas Power Plants
• Demystifying the Mechanisms Behind Emergent Exploration in Goal-conditioned RL
• Learning Wireless Interference Patterns: Decoupled GNN for Throughput Prediction in Heterogeneous Multi-Hop p-CSMA Networks
• On Evaluating Loss Functions for Stock Ranking: An Empirical Analysis With Transformer Model
• Data Understanding Survey: Pursuing Improved Dataset Characterization Via Tensor-based Methods
• Optimal Control Theoretic Neural Optimizer: From Backpropagation to Dynamic Programming
• Contrastive Diffusion Alignment: Learning Structured Latents for Controllable Generation
• Incentive-Based Federated Learning
• Spectral Analysis of Molecular Kernels: When Richer Features Do Not Guarantee Better Generalization
• When Flatness Does (Not) Guarantee Adversarial Robustness
• A Physics Prior-Guided Dual-Stream Attention Network for Motion Prediction of Elastic Bragg Breakwaters
• Generalist vs Specialist Time Series Foundation Models: Investigating Potential Emergent Behaviors in Assessing Human Health Using PPG Signals
• Nonparametric Data Attribution for Diffusion Models
• Stable Prediction of Adverse Events in Medical Time-Series Data
• Enhancing Time-Series Anomaly Detection by Integrating Spectral-Residual Bottom-Up Attention with Reservoir Computing
• Active Measuring in Reinforcement Learning With Delayed Negative Effects
• LLM-ERM: Sample-Efficient Program Learning via LLM-Guided Search
• DARTS-GT: Differentiable Architecture Search for Graph Transformers with Quantifiable Instance-Specific Interpretability Analysis
• Jet Functors and Weil Algebras in Automatic Differentiation: A Geometric Analysis
• SHaRe-SSM: An Oscillatory Spiking Neural Network for Target Variable Modeling in Long Sequences
• Revisit Modality Imbalance at the Decision Layer
• Interaction Concordance Index: Performance Evaluation for Interaction Prediction Methods
• MergeMoE: Efficient Compression of MoE Models via Expert Output Merging
• Towards geological inference with process-based and deep generative modeling, part 1: training on fluvial deposits
• Coder as Editor: Code-driven Interpretable Molecular Optimization
• Learning to Undo: Rollback-Augmented Reinforcement Learning with Reversibility Signals
• Enhancing Time Series Forecasting through Selective Representation Spaces: A Patch Perspective
• On the Identifiability of Tensor Ranks via Prior Predictive Matching
• MX+: Pushing the Limits of Microscaling Formats for Efficient Large Language Model Serving
• Redundancy-Aware Test-Time Graph Out-of-Distribution Detection
• State-Space Models for Tabular Prior-Data Fitted Networks
• Matcha: Multi-Stage Riemannian Flow Matching for Accurate and Physically Valid Molecular Docking
• Multimodal RAG for Unstructured Data:Leveraging Modality-Aware Knowledge Graphs with Hybrid Retrieval
• First Attentions Last: Better Exploiting First Attentions for Efficient Transformer Training
• Geometric Moment Alignment for Domain Adaptation via Siegel Embeddings
• Online Reliable Anomaly Detection via Neuromorphic Sensing and Communications
• Tawa: Automatic Warp Specialization for Modern GPUs with Asynchronous References
• The Pursuit of Diversity: Multi-Objective Testing of Deep Reinforcement Learning Agents
• Causal Discovery for Linear DAGs with Dependent Latent Variables via Higher-order Cumulants
• Active Jammer Localization via Acquisition-Aware Path Planning
• Rethinking Hebbian Principle: Low-Dimensional Structural Projection for Unsupervised Learning
• Efficient Dynamic Structured Sparse Training with Learned Shuffles
• Tackling Time-Series Forecasting Generalization via Mitigating Concept Drift
• Programmatic Representation Learning with Language Models
• To Infinity and Beyond: Tool-Use Unlocks Length Generalization in State Space Models
• Intelligent Dynamic Handover via AI-assisted Signal Quality Prediction in 6G Multi-RAT Networks
• Reinforcement Learning with Stochastic Reward Machines
• Provable Unlearning with Gradient Ascent on Two-Layer ReLU Neural Networks
• Backdoor Unlearning by Linear Task Decomposition
• Efficient Parallel Samplers for Recurrent-Depth Models and Their Connection to Diffusion Language Models
• Identity-Link IRT for Label-Free LLM Evaluation: Preserving Additivity in TVD-MI Scores
• Biology-informed neural networks learn nonlinear representations from omics data to improve genomic prediction and interpretability
• Joint Active RIS Configuration and User Power Control for Localization: A Neuroevolution-Based Approach
• Hybrid Deep Learning Approaches for Classifying Autism from Brain MRI
• Language steering in latent space to mitigate unintended code-switching
• EvoEdit: Evolving Null-space Alignment for Robust and Efficient Knowledge Editing
• R2T: Rule-Encoded Loss Functions for Low-Resource Sequence Tagging
• An Overview of the JPEG AI Learning-Based Image Coding Standard
• DeepMartingale: Duality of the Optimal Stopping Problem with Expressivity
• Post-surgical Endometriosis Segmentation in Laparoscopic Videos
• Switchboard-Affect: Emotion Perception Labels from Conversational Speech
• Long-Term Spatio-Temporal Forecasting of Monthly Rainfall in West Bengal Using Ensemble Learning Approaches
• Classifying and Addressing the Diversity of Errors in Retrieval-Augmented Generation Systems
• Signature in Code Backdoor Detection, how far are we?
• Dynamic SBI: Round-free Sequential Simulation-Based Inference with Adaptive Datasets
• PIShield: Detecting Prompt Injection Attacks via Intrinsic LLM Features
• Exact Dynamics of Multi-class Stochastic Gradient Descent
• deFOREST: Fusing Optical and Radar satellite data for Enhanced Sensing of Tree-loss
• David vs. Goliath: A comparative study of different-sized LLMs for code generation in the domain of automotive scenario generation
• High-Dimensional BWDM: A Robust Nonparametric Clustering Validation Index for Large-Scale Data
• PoissonNet: A Local-Global Approach for Learning on Surfaces
• A novel Information-Driven Strategy for Optimal Regression Assessment
• Flip-Flop Consistency: Unsupervised Training for Robustness to Prompt Perturbations in LLMs
• Constraint-Driven Small Language Models Based on Agent and OpenAlex Knowledge Graph: Mining Conceptual Pathways and Discovering Innovation Points in Academic Papers
• PluriHop: Exhaustive, Recall-Sensitive QA over Distractor-Rich Corpora
• BoardVision: Deployment-ready and Robust Motherboard Defect Detection with YOLO+Faster-RCNN Ensemble
• Low Power Vision Transformer Accelerator with Hardware-Aware Pruning and Optimized Dataflow
• Personalized federated learning, Row-wise fusion regularization, Multivariate modeling, Sparse estimation
• Noise Projection: Closing the Prompt-Agnostic Gap Behind Text-to-Image Misalignment in Diffusion Models
• A Deep State-Space Model Compression Method using Upper Bound on Output Error
• Parameter Identification for Partial Differential Equation with Jump Discontinuities in Coefficients by Markov Switching Model and Physics-Informed Machine Learning
• Decorrelation Speeds Up Vision Transformers
• Response to Discussions of "Causal and Counterfactual Views of Missing Data Models"
• MCbiF: Measuring Topological Autocorrelation in Multiscale Clusterings via 2-Parameter Persistent Homology
• Fast and Scalable Score-Based Kernel Calibration Tests
• Leveraging Code Cohesion Analysis to Identify Source Code Supply Chain Attacks
• Unifying Environment Perception and Route Choice Modeling for Trajectory Representation Learning
• A Geometric Approach to Optimal Experimental Design
• A Multi-Task Deep Learning Framework for Skin Lesion Classification, ABCDE Feature Quantification, and Evolution Simulation
• From Loop Nests to Silicon: Mapping AI Workloads onto AMD NPUs with MLIR-AIR
• Prediction-Specific Design of Learning-Augmented Algorithms
• Secure Sparse Matrix Multiplications and their Applications to Privacy-Preserving Machine Learning
• Learnable Mixed Nash Equilibria are Collectively Rational
• Instruction Set Migration at Warehouse Scale
• VT-Refine: Learning Bimanual Assembly with Visuo-Tactile Feedback via Simulation Fine-Tunin
• DialectGen: Benchmarking and Improving Dialect Robustness in Multimodal Generation
• Learning an Image Editing Model without Image Editing Pairs
• Lost in the Averages: A New Specific Setup to Evaluate Membership Inference Attacks Against Machine Learning Models
• Kernel Neural Operators (KNOs) for Scalable, Memory-efficient, Geometrically-flexible Operator Learning
• Boosting Graph Foundation Model from Structural Perspective
• Hopfield-Fenchel-Young Networks: A Unified Framework for Associative Memory Retrieval
• REX: Causal discovery based on machine learning and explainability techniques
• Exploring the Noise Robustness of Online Conformal Prediction
• Spend Wisely: Maximizing Post-Training Gains in Iterative Synthetic Data Bootstrapping
• FairBatching: Fairness-Aware Batch Formation for LLM Inference
• MedTrust-RAG: Evidence Verification and Trust Alignment for Biomedical Question Answering
• The Role of Social Learning and Collective Norm Formation in Fostering Cooperation in LLM Multi-Agent Systems
• Instructions are all you need: Self-supervised Reinforcement Learning for Instruction Following
• Big Data Approaches to Bovine Bioacoustics: A FAIR-Compliant Dataset and Scalable ML Framework for Precision Livestock Welfare
• A Free Lunch in LLM Compression: Revisiting Retraining after Pruning
• Feature Selection and Regularization in Multi-Class Classification: An Empirical Study of One-vs-Rest Logistic Regression with Gradient Descent Optimization and L1 Sparsity Constraints
• Towards Adaptable Humanoid Control via Adaptive Motion Tracking
• Holdout-Loss-Based Data Selection for LLM Finetuning via In-Context Learning
• LiRA: Linguistic Robust Anchoring for Cross-lingual Large Language Models
• Stealthy Dual-Trigger Backdoors: Attacking Prompt Tuning in LM-Empowered Graph Foundation Models
• Semantic representations emerge in biologically inspired ensembles of cross-supervising neural networks
• From Guess2Graph: When and How Can Unreliable Experts Safely Boost Causal Discovery in Finite Samples?
• E2Edev: Benchmarking Large Language Models in End-to-End Software Development Task
• State Your Intention to Steer Your Attention: An AI Assistant for Intentional Digital Living
• Real-Time Surgical Instrument Defect Detection via Non-Destructive Testing
• Agentic Entropy-Balanced Policy Optimization
• Selective Labeling with False Discovery Rate Control
• Local Causal Discovery for Statistically Efficient Causal Inference
• STANCE: Motion Coherent Video Generation Via Sparse-to-Dense Anchored Encoding
• Just-In-Time Objectives: A General Approach for Specialized AI Interactions
• Knowledge-based Visual Question Answer with Multimodal Processing, Retrieval and Filtering
• An Active Inference Model of Mouse Point-and-Click Behaviour
• Beyond Correctness: Evaluating Subjective Writing Preferences Across Cultures
• Code-driven Number Sequence Calculation: Enhancing the inductive Reasoning Abilities of Large Language Models
• LeapFactual: Reliable Visual Counterfactual Explanation Using Conditional Flow Matching
• GemiRec: Interest Quantization and Generation for Multi-Interest Recommendation
• RLAIF-SPA: Optimizing LLM-based Emotional Speech Synthesis via RLAIF
• Causality Enhancement for Cross-Domain Recommendation
• The Bidding Games: Reinforcement Learning for MEV Extraction on Polygon Blockchain
• In-Context Learning with Unpaired Clips for Instruction-based Video Editing
• Galaxy Morphology Classification with Counterfactual Explanation
• An Efficient Rubric-based Generative Verifier for Search-Augmented LLMs
• When Planners Meet Reality: How Learned, Reactive Traffic Agents Shift nuPlan Benchmarks
• xLLM Technical Report
• FedPPA: Progressive Parameter Alignment for Personalized Federated Learning
• Where are the Whales: A Human-in-the-loop Detection Method for Identifying Whales in High-resolution Satellite Imagery
• Camera Movement Classification in Historical Footage: A Comparative Study of Deep Video Models
• Seesaw: Accelerating Training by Balancing Learning Rate and Batch Size Scheduling
• DEXTER: Diffusion-Guided EXplanations with TExtual Reasoning for Vision Models
• Beyond Multi-Token Prediction: Pretraining LLMs with Future Summaries
• COIG-Writer: A High-Quality Dataset for Chinese Creative Writing with Thought Processes
• Inpainting the Red Planet: Diffusion Models for the Reconstruction of Martian Environments in Virtual Reality
• Finding Answers in Thought Matters: Revisiting Evaluation on Large Language Models with Reasoning
• Cross-Scenario Unified Modeling of User Interests at Billion Scale
• Morphology-Aware Prognostic model for Five-Year Survival Prediction in Colorectal Cancer from H&E Whole Slide Images
• Scaling Artificial Intelligence for Multi-Tumor Early Detection with More Reports, Fewer Masks
• RL-100: Performant Robotic Manipulation with Real-World Reinforcement Learning
• Benchmarking Multimodal Large Language Models for Face Recognition
• Predicting kernel regression learning curves from only raw data statistics
• Learning When Not to Learn: Risk-Sensitive Abstention in Bandits with Unbounded Rewards
• Detecting Early and Implicit Suicidal Ideation via Longitudinal and Information Environment Signals on Social Media
• Reasoning with Sampling: Your Base Model is Smarter Than You Think
• MaskCaptioner : Learning to Jointly Segment and Caption Object Trajectories in Videos
• Predicting Task Performance with Context-aware Scaling Laws
• Circuit Insights: Towards Interpretability Beyond Activations
• LaSeR: Reinforcement Learning with Last-Token Self-Rewarding
• MetaBench: A Multi-task Benchmark for Assessing LLMs in Metabolomics
• Architecture Is All You Need: Diversity-Enabled Sweet Spots for Robust Humanoid Locomotion
• RealDPO: Real or Not Real, that is the Preference
• CBF-RL: Safety Filtering Reinforcement Learning in Training with Control Barrier Functions
• C4D: 4D Made from 3D through Dual Correspondences
• Information Gain-based Policy Optimization: A Simple and Effective Approach for Multi-Turn LLM Agents
• RDD: Retrieval-Based Demonstration Decomposer for Planner Alignment in Long-Horizon Tasks
• LLMs as Scalable, General-Purpose Simulators For Evolving Digital Agent Training
• TokDrift: When LLM Speaks in Subwords but Code Speaks in Grammar
• Attention Is All You Need for KV Cache in Diffusion LLMs
• pi-Flow: Policy-Based Few-Step Generation via Imitation Distillation
• WithAnyone: Towards Controllable and ID Consistent Image Generation
• Terra: Explorable Native 3D World Model with Point Latents
• From Pixels to Words -- Towards Native Vision-Language Primitives at Scale
• Coupled Diffusion Sampling for Training-Free Multi-View Image Editing
• Domain-Independent Dynamic Programming
• TriQXNet: Forecasting Dst Index from Solar Wind Data Using an Interpretable Parallel Classical-Quantum Framework with Uncertainty Quantification
• DELE: Deductive $\mathcal{EL}^{++}$ Embeddings for Knowledge Base Completion
• Robust Counterfactual Inference in Markov Decision Processes
• PoE-World: Compositional World Modeling with Products of Programmatic Experts
• MAFA: A multi-agent framework for annotation
• EMAC+: Embodied Multimodal Agent for Collaborative Planning with VLM+LLM
• Plugging Schema Graph into Multi-Table QA: A Human-Guided Framework for Reducing LLM Reliance
• Interpretability and Transparency-Driven Detection and Transformation of Textual Adversarial Examples (IT-DT)
• Quantum Polar Metric Learning: Efficient Classically Learned Quantum Embeddings
• Natural Language Processing RELIES on Linguistics
• Why do explanations fail? A typology and discussion on failures in XAI
• Ada-KV: Optimizing KV Cache Eviction by Adaptive Budget Allocation for Efficient LLM Inference
• Visual Stereotypes of Autism Spectrum in Janus-Pro-7B, DALL-E, Stable Diffusion, SDXL, FLUX, and Midjourney
• Say My Name: a Model's Bias Discovery Framework
• The Fluorescent Veil: A Stealthy and Effective Physical Adversarial Patch Against Traffic Sign Recognition
• MIO: A Foundation Model on Multimodal Tokens
• SFTMix: Elevating Language Model Instruction Tuning with Mixup Recipe
• Emergent Visual Grounding in Large Multimodal Models Without Grounding Supervision
• VoxelPrompt: A Vision Agent for End-to-End Medical Image Analysis
• CoreGuard: Safeguarding Foundational Capabilities of LLMs Against Model Stealing in Edge Deployment
• AI-generated Essays: Characteristics and Implications on Automated Scoring and Academic Integrity
• Disentangled and Self-Explainable Node Representation Learning
• An AI-Driven Multimodal Smart Home Platform for Continuous Monitoring and Assistance in Post-Stroke Motor Impairment
• Moto: Latent Motion Token as the Bridging Language for Learning Robot Manipulation from Videos
• VERITAS: Verifying the Performance of AI-native Transceiver Actions in Base-Stations
• The Last Dependency Crusade: Solving Python Dependency Conflicts with LLMs
• FedRTS: Federated Robust Pruning via Combinatorial Thompson Sampling
• The simulation of judgment in LLMs
• Evaluating Sakana's AI Scientist: Bold Claims, Mixed Results, and a Promising Future?
• Never too Prim to Swim: An LLM-Enhanced RL-based Adaptive S-Surface Controller for AUVs under Extreme Sea Conditions
• A Neural Symbolic Model for Space Physics
• Overcoming Sparsity Artifacts in Crosscoders to Interpret Chat-Tuning
• EDIT: Enhancing Vision Transformers by Mitigating Attention Sink through an Encoder-Decoder Architecture
• Benchmarking Adversarial Robustness to Bias Elicitation in Large Language Models: Scalable Automated Assessment with LLM-as-a-Judge
• On Equivariance and Fast Sampling in Video Diffusion Models Trained with Warped Noise
• Restoring Calibration for Aligned Large Language Models: A Calibration-Aware Fine-Tuning Approach
• Absolute Zero: Reinforced Self-play Reasoning with Zero Data
• LLMs' Suitability for Network Security: A Case Study of STRIDE Threat Modeling
• The ML.ENERGY Benchmark: Toward Automated Inference Energy Measurement and Optimization
• Internet of Agents: Fundamentals, Applications, and Challenges
• ConDiSim: Conditional Diffusion Models for Simulation Based Inference
• APEX: Empowering LLMs with Physics-Based Task Planning for Real-time Insight
• Checkpoint-GCG: Auditing and Attacking Fine-Tuning-Based Prompt Injection Defenses
• KL-regularization Itself is Differentially Private in Bandits and RLHF
• Thinker: Learning to Think Fast and Slow
• Adaptive Budget Allocation for Orthogonal-Subspace Adapter Tuning in LLMs Continual Learning
• ToolPRM: Fine-Grained Inference Scaling of Structured Outputs for Function Calling
• SimKO: Simple Pass@K Policy Optimization
• Agentic NL2SQL to Reduce Computational Costs
• RoboGPT-R1: Enhancing Robot Planning with Reinforcement Learning
• Boosting Instruction Following at Scale
• Where to Search: Measure the Prior-Structured Search Space of LLM Agents
• LabOS: The AI-XR Co-Scientist That Sees and Works With Humans
• The Gatekeeper Knows Enough
• Mapping Smarter, Not Harder: A Test-Time Reinforcement Learning Agent That Improves Without Labels or Model Updates
• Budget-aware Test-time Scaling via Discriminative Verification
• TRI-DEP: A Trimodal Comparative Study for Depression Detection Using Speech, Text, and EEG
• Stable but Miscalibrated: A Kantian View on Overconfidence from Filters to Large Language Models
• GroundedPRM: Tree-Guided and Fidelity-Aware Process Reward Modeling for Step-Level Reasoning
• Agentic Design of Compositional Machines
• Generative AI in Heritage Practice: Improving the Accessibility of Heritage Guidance
• Reversing the Lens: Using Explainable AI to Understand Human Expertise
• GQVis: A Dataset of Genomics Data Questions and Visualizations for Generative AI
• Leveraging Wireless Sensor Networks for Real-Time Monitoring and Control of Industrial Environments
• A2AS: Agentic AI Runtime Security and Self-Defense
• Towards Neurocognitive-Inspired Intelligence: From AI's Structural Mimicry to Human-Like Functional Cognition
• Bridging the Semantic Gap: Contrastive Rewards for Multilingual Text-to-SQL
• A Linguistics-Aware LLM Watermarking via Syntactic Predictability
• Users as Annotators: LLM Preference Learning from Comparison Mode
• Informed Routing in LLMs: Smarter Token-Level Computation for Faster Inference
• Entropy Meets Importance: A Unified Head Importance-Entropy Score for Stable and Efficient Transformer Pruning
• ConDABench: Interactive Evaluation of Language Models for Data Analysis
• SIMBA UQ: Similarity-Based Aggregation for Uncertainty Quantification in Large Language Models
• Seeing Hate Differently: Hate Subspace Modeling for Culture-Aware Hate Speech Detection
• Meronymic Ontology Extraction via Large Language Models
• ADMIT: Few-shot Knowledge Poisoning Attacks on RAG-based Fact Checking
• Serialized EHR make for good text representations
• Information flow in multilayer perceptrons: an in-depth analysis
• DynaSpec: Context-aware Dynamic Speculative Sampling for Large-Vocabulary Language Models
• On-device System of Compositional Multi-tasking in Large Language Models
• Revisiting the UID Hypothesis in LLM Reasoning Traces
• ConsistencyAI: A Benchmark to Assess LLMs' Factual Consistency When Responding to Different Demographic Groups
• BenchPress: A Human-in-the-Loop Annotation System for Rapid Text-to-SQL Benchmark Curation
• Harnessing Consistency for Robust Test-Time LLM Ensemble
• Multimodal Retrieval-Augmented Generation with Large Language Models for Medical VQA
• From Craft to Constitution: A Governance-First Paradigm for Principled Agent Engineering
• Benchmarking Correctness and Security in Multi-Turn Code Generation
• ShishuLM: Lightweight Language Model with Hybrid Decoder-MLP Architecture and Paired Weight Sharing
• Ensembling Large Language Models to Characterize Affective Dynamics in Student-AI Tutor Dialogues
• Self-Training with Dynamic Weighting for Robust Gradual Domain Adaptation
• Deep Edge Filter: Return of the Human-Crafted Layer in Deep Learning
• FFT-Accelerated Auxiliary Variable MCMC for Fermionic Lattice Models: A Determinant-Free Approach with $O(N\log N)$ Complexity
• CoLoR-GAN: Continual Few-Shot Learning with Low-Rank Adaptation in Generative Adversarial Networks
• Unlocking the Potential of Diffusion Language Models through Template Infilling
• Joint Discriminative-Generative Modeling via Dual Adversarial Training
• FRACCO: A gold-standard annotated corpus of oncological entities with ICD-O-3.1 normalisation
• What Layers When: Learning to Skip Compute in LLMs with Residual Gates
• Catch Your Breath: Adaptive Computation for Self-Paced Sequence Production
• PAGE: Prompt Augmentation for text Generation Enhancement
• Order from Chaos: Comparative Study of Ten Leading LLMs on Unstructured Data Categorization
• Physics-Informed autoencoder for DSC-MRI Perfusion post-processing: application to glioma grading
• Incomplete Multi-view Clustering via Hierarchical Semantic Alignment and Cooperative Completion
• Reliable Fine-Grained Evaluation of Natural Language Math Proofs
• A Survey on Collaborating Small and Large Language Models for Performance, Cost-effectiveness, Cloud-edge Privacy, and Trustworthiness
• K-frames: Scene-Driven Any-k Keyframe Selection for long video understanding
• Guarding the Guardrails: A Taxonomy-Driven Approach to Jailbreak Detection
• Bayes or Heisenberg: Who(se) Rules?
• GenCellAgent: Generalizable, Training-Free Cellular Image Segmentation via Large Language Model Agents
• Dual-attention ResNet outperforms transformers in HER2 prediction on DCE-MRI
• Narrow Finetuning Leaves Clearly Readable Traces in Activation Differences
• Benefits and Limitations of Communication in Multi-Agent Reasoning
• Schema for In-Context Learning
• Knowledge Reasoning Language Model: Unifying Knowledge and Language for Inductive Knowledge Graph Reasoning
• AI Debaters are More Persuasive when Arguing in Alignment with Their Own Beliefs
• Synthesizing Agentic Data for Web Agents with Progressive Difficulty Enhancement Mechanisms
• Readability $\ne$ Learnability: Rethinking the Role of Simplicity in Training Small Language Models
• LLMs Can Get "Brain Rot"!
• Big Reasoning with Small Models: Instruction Retrieval at Inference Time
• Less is More: Improving LLM Reasoning with Minimal Test-Time Intervention
• Static Sandboxes Are Inadequate: Modeling Societal Complexity Requires Open-Ended Co-Evolution in LLM-Based Multi-Agent Simulations
• Efficient Few-Shot Learning in Remote Sensing: Fusing Vision and Vision-Language Models
• Finding Holes: Pathologist Level Performance Using AI for Cribriform Morphology Detection in Prostate Cancer
• REAP the Experts: Why Pruning Prevails for One-Shot MoE compression
• Conditional Clifford-Steerable CNNs with Complete Kernel Basis for PDE Modeling
• Context-Selective State Space Models: Feedback is All You Need
• Think Globally, Group Locally: Evaluating LLMs Using Multi-Lingual Word Grouping Games
• One Bug, Hundreds Behind: LLMs for Large-Scale Bug Discovery
• Cyber-Resilient System Identification for Power Grid through Bayesian Integration
• Optical Computation-in-Communication enables low-latency, high-fidelity perception in telesurgery
• On the expressivity of sparse maxout networks
• Exploratory Causal Inference in SAEnce
• DiffOPF: Diffusion Solver for Optimal Power Flow
• Every Language Model Has a Forgery-Resistant Signature
• Unlocking Out-of-Distribution Generalization in Transformers via Recursive Latent Space Reasoning
• Extracting latent representations from X-ray spectra. Classification, regression, and accretion signatures of Chandra sources
• Toward Cybersecurity-Expert Small Language Models
• Inferred global dense residue transition graphs from primary structure sequences enable protein interaction prediction via directed graph convolutional neural networks
• FinAI Data Assistant: LLM-based Financial Database Query Processing with the OpenAI Function Calling API
• Towards Reversible Model Merging For Low-rank Weights
• Virtually Being: Customizing Camera-Controllable Video Diffusion Models with Multi-View Performance Captures
• MAFA: A Multi-Agent Framework for Enterprise-Scale Annotation with Configurable Task Adaptation
• DPRF: A Generalizable Dynamic Persona Refinement Framework for Optimizing Behavior Alignment Between Personalized LLM Role-Playing Agents and Humans
• LiteStage: Latency-aware Layer Skipping for Multi-stage Reasoning
• Large Scale Retrieval for the LinkedIn Feed using Causal Language Models
• Scaling Test-Time Compute to Achieve IOI Gold Medal with Open-Weight Models
• Spatial Computing Communications for Multi-User Virtual Reality in Distributed Mobile Edge Computing Network
• Reinforcement Learning for Unsupervised Domain Adaptation in Spatio-Temporal Echocardiography Segmentation
• Policy Regularized Distributionally Robust Markov Decision Processes with Linear Function Approximation
• Do Joint Language-Audio Embeddings Encode Perceptual Timbre Semantics?
• CAST: Compositional Analysis via Spectral Tracking for Understanding Transformer Layer Functions
• Less is More: Denoising Knowledge Graphs For Retrieval Augmented Generation
• PRISM: Agentic Retrieval with LLMs for Multi-Hop Question Answering
• Beyond a Single Perspective: Towards a Realistic Evaluation of Website Fingerprinting Attacks
• Learning Human-Humanoid Coordination for Collaborative Object Carrying
• TED++: Submanifold-Aware Backdoor Detection via Layerwise Tubular-Neighbourhood Screening
• Expertise need not monopolize: Action-Specialized Mixture of Experts for Vision-Language-Action Learning
• Watermarking for Factuality: Guiding Vision-Language Models Toward Truth via Tri-layer Contrastive Decoding
• MERLIN: A Testbed for Multilingual Multimodal Entity Recognition and Linking
• Column Generation Using Domain-Independent Dynamic Programming
• Evaluating & Reducing Deceptive Dialogue From Language Models with Multi-turn RL
• A Robust Classification Method using Hybrid Word Embedding for Early Diagnosis of Alzheimer's Disease
• Stop-RAG: Value-Based Retrieval Control for Iterative RAG
• A Density-Informed Multimodal Artificial Intelligence Framework for Improving Breast Cancer Detection Across All Breast Densities
• BinCtx: Multi-Modal Representation Learning for Robust Android App Behavior Detection
• Beyond One World: Benchmarking Super Heros in Role-Playing Across Multiversal Contexts
• CURE: Confidence-driven Unified Reasoning Ensemble Framework for Medical Question Answering
• SUM-AgriVLN: Spatial Understanding Memory for Agricultural Vision-and-Language Navigation
• From Binary to Bilingual: How the National Weather Service is Using Artificial Intelligence to Develop a Comprehensive Translation Program
• Are My Optimized Prompts Compromised? Exploring Vulnerabilities of LLM-based Optimizers
• Beat Detection as Object Detection
• Decision Oriented Technique (DOTechnique): Finding Model Validity Through Decision-Maker Context
• Do Slides Help? Multi-modal Context for Automatic Transcription of Conference Talks
• Do Large Language Models Show Biases in Causal Learning? Insights from Contingency Judgment
• GammaZero: Learning To Guide POMDP Belief Space Search With Graph Representations
• Position: Require Frontier AI Labs To Release Small "Analog" Models
• Generating Fair Consensus Statements with Social Choice on Token-Level MDPs
• STEMS: Spatial-Temporal Enhanced Safe Multi-Agent Coordination for Building Energy Management
• Formalizing the Safety, Security, and Functional Properties of Agentic AI Systems
• A Multimodal Approach to Heritage Preservation in the Context of Climate Change
• CodeEvolve: An open source evolutionary coding agent for algorithm discovery and optimization
• Combining Reinforcement Learning and Behavior Trees for NPCs in Video Games with AMD Schola
• JEDA: Query-Free Clinical Order Search from Ambient Dialogues
• ARM-FM: Automated Reward Machines via Foundation Models for Compositional Reinforcement Learning
• Implementation of AI in Precision Medicine
• Echoes of Human Malice in Agents: Benchmarking LLMs for Multi-Turn Online Harassment Attacks
• LiveResearchBench: A Live Benchmark for User-Centric Deep Research in the Wild
• Towards Agentic Self-Learning LLMs in Search Environment
• MorphoBench: A Benchmark with Difficulty Adaptive to Model Reasoning
• A Guardrail for Safety Preservation: When Safety-Sensitive Subspace Meets Harmful-Resistant Null-Space
• Terrarium: Revisiting the Blackboard for Multi-Agent Safety, Privacy, and Security Studies
• Metacognitive Self-Correction for Multi-Agent System via Prototype-Guided Next-Execution Reconstruction
• AI for Service: Proactive Assistance with AI Glasses
• Can MLLMs Absorb Math Reasoning Abilities from LLMs as Free Lunch?
• Hi-Agent: Hierarchical Vision-Language Agents for Mobile Device Control
• IMAGINE: Integrating Multi-Agent System into One Model for Complex Reasoning and Planning
• Eliminating Negative Occurrences of Derived Predicates from PDDL Axioms
• Helmsman: Autonomous Synthesis of Federated Learning Systems via Multi-Agent Collaboration
• JSPLIT: A Taxonomy-based Solution for Prompt Bloating in Model Context Protocol
• Symbol Grounding in Neuro-Symbolic AI: A Gentle Introduction to Reasoning Shortcuts
• LLM Agents Beyond Utility: An Open-Ended Perspective
• ColorBench: Benchmarking Mobile Agents with Graph-Structured Framework for Complex Long-Horizon Tasks
• Beyond Hallucinations: The Illusion of Understanding in Large Language Models
• Machine Learning and Public Health: Identifying and Mitigating Algorithmic Bias through a Systematic Review
• TITAN: Graph-Executable Reasoning for Cyber Threat Intelligence
• NAEL: Non-Anthropocentric Ethical Logic
• Practical, Utilitarian Algorithm Configuration
• Purifying Task Vectors in Knowledge-Aware Subspace for Model Merging
• Cognitive-Aligned Spatio-Temporal Large Language Models For Next Point-of-Interest Prediction