Papers

New on arXiv

Sat 23 May 2026

263 papers

Geometry-Aware Tabular DiffusionDiffusion models → Flow matching, time series and science
Graph Mamba Survival Analysis Based on Topology-Aware orderingState-space sequence models (S4, Mamba) → Mamba for medical image analysis
Silent Failures in Physical AI: A Literature Review of Runtime Action Authorization for Autonomous SystemsWorld models → World models for navigation +1
From Frontier to Shadow AI: A Simmering Threat to Assurance and Security in Critical InfrastructureLLM agents and tool use → Agent threats, guardrails and governance
RAGe: A Retrieval-Augmented Generation Evaluation FrameworkRetrieval-augmented generation → RAG pipeline design and evaluation
A Systematic Evaluation of Retrieval-Augmented Generation and Language Models for Space OperationsRetrieval-augmented generation → RAG pipeline design and evaluation
PRIMA: Operational Patterns for Resilient Multi-Agent Research with Verifiable Identity and Convergent FeedbackLLM agents and tool use → Software engineering coding agents
From Theory to Decision Rule: Calibrating the Noisy-Label Crossover for Vision-Language Model Weak Supervision Across Three Medical-Imaging BenchmarksMultimodal LLMs / vision-language models → Medical vision-language models
Drift-Resistant Navigation World Model with Anchored Epipolar GuidanceWorld models → World models for navigation
A Contractive Feedback Semantics for Reinforcement LearningDeep reinforcement learning methods → RL theory and sample complexity
Proper Scoring Rules for Agentic Uncertainty QuantificationLLM agents and tool use → General LLM agent frameworks and benchmarks
Automated Detection and Classification of Delusion-related Content in Naturalistic Audio Diaries Using Multi-Agent Language ModelsLLM agents and tool use → Medical and scientific domain agents
How Neural Reward Models Learn Features for Policy Optimization: A Single-Index AnalysisLLM alignment / RLHF / preference optimisation → Reward modeling and RLHF
Reinforcement Learning for Reachability: Guaranteeing Asymptotic OptimalityDeep reinforcement learning methods → RL theory and sample complexity
Streaming Reinforcement Learning under Partial Observability with Real-Time Recurrent LearningDeep reinforcement learning methods → Deep RL algorithms and representations
MDIA: A Multi-Agent Diagnostic Intelligence Pipeline on HealthBench ProfessionalLLM agents and tool use → Medical and scientific domain agents
The Path Matters: Learning a Token-Commitment Policy for Diffusion Language ModelsDiffusion models → Discrete and text diffusion
CP-Agent: A Calibrated Risk-Controlled Agent for Feedback-Driven Competitive ProgrammingLLM agents and tool use → General LLM agent frameworks and benchmarks
Sum of Costs Diffusion with Dynamic Guidance for Motion PlanningDiffusion models → Diffusion and flow policies for robotics
HoloFair: Unified T2I Fairness Evaluation and Fair-GRPO DebiasingLLM reasoning → Multimodal reasoning
Mix-MoE: Improving Multilingual Machine Translation of Large Language Models through Mixed MoEsMixture-of-experts → MoE large language models
Reasoning to Align: Implicit Reasoning in Diffusion Transformers for Video EditingLLM reasoning → Multimodal reasoning +1
Measuring Reasoning Quality in LLMs: A Multi-Dimensional Behavioral FrameworkLLM reasoning → Reasoning benchmarks and evaluation
How Many Tools Should an LLM Agent See? A Chance-Corrected AnswerLLM agents and tool use → Tool learning and function calling
IterInject: Indirect Prompt Injection Against LLM Agents via Feedback-Guided Iterative OptimizationLLM agents and tool use → Agent threats, guardrails and governance +1
AVBench: Human-Aligned and Automated Evaluation Benchmark for Audio-Video Generative ModelsLLM alignment / RLHF / preference optimisation → Diffusion and multimodal model alignment
Towards Low-Gravity Planetary Exploration using Reinforcement Learning for Walking, Jumping, and In-flight Attitude ControlDeep reinforcement learning methods → RL for robot locomotion and manipulation
Understanding the Impact of Geometric Foundation Models on Vision-Language-Action ModelsVision-language-action models → 3D and spatial-aware VLA
GlobalDentBench: A Multinational Benchmark for Evaluating LLM Clinical Reasoning in Dentistry with Expert CalibrationLLM reasoning → Clinical and biomedical applications
HiMed: Incentivizing Hindi Reasoning in Medical LLMsLLM reasoning → Clinical and biomedical applications
Beyond Generative Priors: Minority Sampling with JEPA-Guided DiffusionDiffusion models → Diffusion foundations and sampling
DexSIM: Real-time Dexterous Simulation with Unified Causal Video DiffusionDiffusion models → Diffusion and flow policies for robotics
Measuring the Depth of LLM Unlearning via Activation PatchingLLM safety and security → Machine unlearning +1
Guarded Repair for Harm-Aware Post-hoc Replacement of LLM Mathematical ReasoningLLM reasoning → Mathematical reasoning
CSP-Atlas: Concept-Specific Neural Circuits in a Sparse Python TransformerMechanistic interpretability → Circuits and mechanistic analysis
Correcting Visual Blur Induced by Attention Distraction to Reduce Hallucinations: Algorithm and TheoryMultimodal LLMs / vision-language models → Hallucination mitigation in LVLMs
Agent-as-Peer-Debriefer: A Multi-Agent Framework with Perspective-Based Refinement for Qualitative AnalysisLLM agents and tool use → General LLM agent frameworks and benchmarks
Hera: Learning Long-Horizon Coordination for Device-Cloud Collaborative LLM AgentsLLM agents and tool use → General LLM agent frameworks and benchmarks
Learning to Reason Efficiently with A* Post-TrainingLLM reasoning → Test-time compute and efficient thinking
From Replacement to Orchestration: A Socio-Technical Architecture for Agentic AI in Corporate R&DLLM agents and tool use → Software engineering coding agents
World Models as Group ActionsWorld models → Video-generation world models and simulators
Polymorphism Is Rotation: Operational Mechanistic Interpretability from a Two-Layer Transformer to Pythia-70mMechanistic interpretability → Circuits and mechanistic analysis
AstroMind: A High-Fidelity Benchmark for Spacecraft Behavior Reasoning Based on Large Language ModelsLLM reasoning → Knowledge- and tool-augmented reasoning
EMA: Effort Metric Attention for Anatomical Effort-Guided Human Motion DiffusionDiffusion models → Video generation
PEDESTRIANQA: A Benchmark for Vision-Language Models on Pedestrian Intention and Trajectory PredictionMultimodal LLMs / vision-language models → Driving VLMs and open-vocabulary detection
The Multilingual Curse at the Retrieval Layer: Evidence from AmharicRetrieval-augmented generation → Generation and context integration
Ellipsoid Control: A White-list Jailbreak Defense via Benign Latent ModelingLLM safety and security → Jailbreaking
Jailbreak to Protect: Buffering and Reinforcing via Temporary Jailbreaking for Safe Fine-Tuning in Large Language ModelsLLM safety and security → Jailbreaking
PALoRA: Projection-Adaptive LoRA for Preserving Reasoning in Large Language ModelsLLM reasoning → Test-time compute and efficient thinking
Deep ZakaiJ: Structured Filtering for Jump-Diffusion Time Series ForecastingDiffusion models → Flow matching, time series and science
RL with Learnable Textual Feedback: A Bilevel ApproachLLM reasoning → Reinforcement learning for reasoning
Rethinking Federated Unlearning via the Lens of MemorizationLLM safety and security → Federated unlearning and poisoning
Emission-Aware Reinforcement Learning for Sustainable Electric Vehicle Charging and Carbon Dioxide Reduction Under Varying Renewable PenetrationDeep reinforcement learning methods → RL for energy, networks and operations
DemoEvolve: Overcoming Sparse Feedback in Agentic Harness Evolution with DemonstrationsLLM agents and tool use → Software engineering coding agents
Steering Beyond the Support: Adversarial Training on Unsupervised Jailbroken Activation SimulationLLM safety and security → Jailbreaking
NudgeVAD: Language-Nudged End-to-End Driving via FiLM ResidualsEnd-to-end autonomous driving → Reasoning VLA and VLM driving policies
Unveil: Unified Visual-Textual Integration and Distillation for Multi-modal Document RetrievalKnowledge distillation → Multimodal distillation
Hypothesis Generation and Inductive Inference in Children and Language ModelsLLM agents and tool use → General LLM agent frameworks and benchmarks
ECHO: Terminal Agents Learn World Models for FreeWorld models → World models for language agents +2
Phi-Noise: Training-Free Temporal Video Conditioning via Phase-Based Noise ManipulationDiffusion models → Video generation
Reasoning as an Attack Surface: Adaptive Evolutionary CoT Jailbreaks for LLMsLLM reasoning → Reasoning for agents, planning and code +1
Med-R2: An Adversarial Benchmark for Evidence-Grounded Reasoning in Medical VLMsLLM reasoning → Clinical and biomedical applications +1
Market Regime Council for Dynamic Credit Assignment in Multi-Agent LLM Decision SystemsLLM agents and tool use → Financial trading and forecasting agents
AgentFugue: Agent Scaling for Long-Horizon Tasks through Collective ReasoningLLM agents and tool use → General LLM agent frameworks and benchmarks
OmniEgo-R2: A Routed Reasoning Framework for the 1st Cross-Domain EgoCross Challenge at CVPR 2026LLM reasoning → Spatial and embodied reasoning +1
TempRet: Temporal Enhancement and Two-Stage Reranking for CVPR 2026 EPIC-KITCHENS-100 Multi-Instance Retrieval ChallengeMultimodal LLMs / vision-language models → Video understanding with MLLMs
SAM: State-Adaptive Memory for Long-Horizon Reasoning AgentLLM reasoning → Reasoning for agents, planning and code +1
EgoProx: Evaluating MLLMs on Egocentric 3D Proximity Reasoning Across a Cognitive HierarchyLLM reasoning → Spatial and embodied reasoning
Code2UML: Agentic LLMs with context engineering for scalable software visualizationLLM agents and tool use → Software engineering coding agents
Vision-Guided Outdoor Flight and Obstacle Evasion via Reinforcement LearningDeep reinforcement learning methods → RL for robot navigation and UAV control
A Reinforcement Learning Inspired Latent Yield Based Adaptive Algorithm Switching MechanismDeep reinforcement learning methods → RL for games and combinatorial optimization
Smoother Action Chunking Flow Policy via Prior-Corrected Orthogonal Trust-Region GuidanceDiffusion models → Diffusion and flow policies for robotics
SEAL: Synergistic Co-Evolution of Agents and Learning EnvironmentsLLM agents and tool use → General LLM agent frameworks and benchmarks
Benchmarking the Limits of In-Context Reinforcement Learning for Ad-Hoc TeamworkDeep reinforcement learning methods → RL for games and combinatorial optimization
Poisoning the Watchtower: Prompt Injection Attacks Against LLM-Augmented Security Operations Through Adversarial Log ContentLLM safety and security → Prompt injection and agent security
Generative OOD-regularized Model-based Policy OptimizationOffline RL and imitation learning from data → Offline RL algorithms +1
Dual Prototype-Conditioned Diffusion Model for Scalable Multi-Class Unsupervised Anomaly Detection in Large Category SpacesDiffusion models → Flow matching, time series and science
ConceptM3oE: Concept-Guided Multimodal Mixture of Experts for Interpretable Computational PathologyLLM reasoning → Clinical and biomedical applications +1
VectorArk: Learning Practical Image Vectorization with Rounded Polygon RepresentationMultimodal LLMs / vision-language models → Chart understanding and anomaly detection
Understanding and Mitigating Premature Confidence for Better LLM ReasoningLLM reasoning → Test-time compute and efficient thinking
Distilling Game Code World Model Generation into Lightweight Large Language ModelsWorld models → World models for language agents
Structure-Aware RAG: Structured Retrieval Augmented Generation from Noisy Data for Conversational AgentsRetrieval-augmented generation → Conversational and speech RAG +1
Refined Analysis of Entropy-Regularized Actor-CriticDeep reinforcement learning methods → RL theory and sample complexity
SparseWorld: Enhancing End-to-End Autonomous Driving via World Models with Sparse Scene RepresentationWorld models → World models for autonomous driving +1
Partner-Aware Hierarchical Skill Discovery for Robust Human-AI CollaborationDeep reinforcement learning methods → RL training of language and agent models
Evolving Robustness--Exploration Trade-off in Online Reinforcement Learning via Quantile Bayesian Risk MDPsDeep reinforcement learning methods → RL theory and sample complexity
Adaptive Human-AI Coordination via Hierarchical Action DisentanglementDeep reinforcement learning methods → RL training of language and agent models
CurveRL: Principled Distribution-Aware Context Reweighting for LLM ReasoningLLM reasoning → Reinforcement learning for reasoning
Causal Physics Steering in Video World Models via Concept Activation VectorsWorld models → Latent predictive world models
Unified 3D Scene Understanding Through Physical World ModelingWorld models → Video-generation world models and simulators +1
Five Queries Are Enough: Query-Efficient and Surrogate-Free Membership Inference Attacks on RAG via EntailmentRetrieval-augmented generation → Privacy and security of RAG +1
Reframing LLM Agent Security as an Agent-Human Interaction ProblemLLM agents and tool use → Agent threats, guardrails and governance
ChaosBench-Logic v2: Evaluating LLM Logical Reasoning over Dynamical Systems at ScaleLLM reasoning → Test-time compute and efficient thinking
Cross-Modal Action Recognition in Egocentric Video Using Mamba: Integrating RGB and Hand Skeleton Streams via CLS Token Fusion StrategiesState-space sequence models (S4, Mamba) → Video, motion and generative Mamba models
Few-Shot Degradation Is Not What It Seems: Behavioral Evidence, Representation Analysis, and a Random-Text Control Across 12 Models, 2 Tasks, and 2 Architectures
Large Language Models Threaten Double-blind Review
OpenAI Privacy Filter: A Cross-Lingual, Cross-Domain PII Evaluation Across 32 Benchmarks
Procedural Knowledge Is Not Low-Rank: Why LoRA Fails to Internalize Multi-Step Procedures
Benchmarking the Personalization Capabilities of Large Language Models
The Open Source Economic Index of AI Adoption and Capability

All 263