Papers

New on arXiv

Sun 7 Jun 2026

249 papers

WARA: Toward Automated Wireless Optimization Research with Closed-Loop LLM AgentsLLM agents and tool use → Agentic AI for wireless networks
Measuring Cognitive Engagement in Collaborative Discourse with an Extended ICAP Framework: Comparing Human Annotation, In-Context Learning, and Reflective LLM AgentsLLM agents and tool use → General LLM agent frameworks and benchmarks
HeraSys: Collaborative Serving of Multiple LLM Workflows via Fine-Grained End-to-End OptimizationLLM agents and tool use → Software engineering coding agents
X-Tokenizer: A Multimodal Action Tokenizer for Vision-Language-Action PretrainingVision-language-action models → Efficient VLA inference and adaptation
RKSC: Reasoning-Aware KV Cache Sharing and Confident Early Exit for Multi-Step LLM InferenceLLM reasoning → Test-time compute and efficient thinking +1
One Lens, Many Worlds : A Capability-Typed Interface for World-Model InterpretabilityWorld models → World models for language agents +1
GitInject: Real-World Prompt Injection Attacks in AI-Powered CI/CD PipelinesLLM safety and security → Prompt injection and agent security
Sample Where You Struggle: Sharpening Base Model Reasoning via Entropy-Guided Power SamplingLLM reasoning → Test-time compute and efficient thinking
Co-GLANCE: Uncertainty-Aware Active Perception for Heterogeneous Robot TeamingMultimodal LLMs / vision-language models → Embodied and 3D vision-language models
Benchmarking Vision-Language-Action Models on SO-101: Failure and Recovery AnalysisVision-language-action models → Failure recovery and robustness +1
PerspectiveGap: A Benchmark for Multi-Agent Orchestration PromptingLLM agents and tool use → General LLM agent frameworks and benchmarks
Can the Environment Speak for Itself? T2-GRPO: A Turn-Trajectory Group Relative Policy Optimization for Caregiver AgentsLLM reasoning → Clinical and biomedical applications +1
Building Customer Support AI Agents at 100M-User Scale: An Evaluation-Driven FrameworkLLM agents and tool use → General LLM agent frameworks and benchmarks
Generalizing Geometry-Guided Mamba as a Plug-and-Play Context Module for CNN-based Semantic SegmentationState-space sequence models (S4, Mamba) → Vision Mamba backbones
PaperMentor: A Human-Centered Multi-Agent Writing Tutor for AI Research Papers on OverleafLLM agents and tool use → Web, search, and research agents
Intrinsic Selection and Particle Resampling for Inference-Time Scaling Beyond Domain VerifiabilityLLM reasoning → Test-time compute and efficient thinking
ZIPP:Zero-shot Image Personalization from PersonasDiffusion models → Image generation, text-to-image and editing
CSFlow: Aligning Flow Matching with Human Contrast SensitivityDiffusion models → Flow matching, time series and science
Inference-Time Conformal Reasoning with Valid Factuality Control for Large Language ModelsLLM reasoning → Test-time compute and efficient thinking
Knowledge Graphs and Reasoning LLMs for Finding Simple Yet Effective Transcriptomic Perturbation PredictorsLLM reasoning → Clinical and biomedical applications
Momentum for Reasoning: Dense Intrinsic Signals in Policy OptimizationLLM reasoning → Reinforcement learning for reasoning
STAR: Rethinking MoE Routing as Structure-Aware Subspace LearningMixture-of-experts → Expert routing strategies
Continuous Language Diffusion as a Decoder-Interface ProblemDiffusion models → Discrete and text diffusion
The Amplifying Mirror: Locating and Steering the Partisan Direction inside a Large Language ModelMechanistic interpretability → Probing deception, bias and beliefs
MaskAlign: Token-Subset Representation Alignment for Efficient Diffusion TrainingDiffusion models → Diffusion foundations and sampling
DeepMine-Mamba: Mitigating Information Dilution in Mamba-Based State Space Models for Document Image BinarizationState-space sequence models (S4, Mamba) → Mamba for image restoration
Beyond Consistency: Preserving Temporal Structure in Zero-Shot Video EditingDiffusion models → Video generation
How Many Counterfactuals Does It Take? Probing VLM Hallucinations Through Circuits and Causal EffectsMultimodal LLMs / vision-language models → Hallucination mitigation in LVLMs
Unifying Object-Centric World Models and Diffusion Policy: A Hierarchical Framework for Multi-Stage Robotic TasksWorld models → Embodied and robot world models +1
Less Is More: Training-Free Acceleration Framework of 3D Diffusion Models for Low-Count PET Denoising via Global-Local Trajectory ReductionDiffusion models → Medical imaging and segmentation
Guided Discovery of New Behaviors using Diffusion PoliciesDiffusion models → Diffusion and flow policies for robotics
Dream-Tac: A Unified Tactile World Action Model for Contact-Rich Robot ManipulationWorld models → Embodied and robot world models
Structure-Conditioned Actor-Critic Branches for Quality-Diversity Reinforcement LearningDeep reinforcement learning methods → RL theory and sample complexity
Artificial Intelligence for Mathematical Reasoning: An Integrated Survey of Language Models, Neuro-symbolic Systems, and Verified DiscoveryLLM reasoning → Mathematical reasoning
Thinking Without Images: Internalizing Visual Manipulation with On-Policy Self-DistillationLLM reasoning → Multimodal reasoning +1
Structuring agentic AI for HPC code modernizationLLM agents and tool use → Software engineering coding agents
PRPO: Perception-Reinforced Policy Optimization via Token-Level Dynamic Advantage ReshapingLLM reasoning → Multimodal reasoning +1
ConMem: Structured Memory-Guided Adaptation in Training-Free Multi-Agent SystemsLLM agents and tool use → Agent memory systems
Agentic Search for Counterfactual Recourse under Fixed LLM BudgetsLLM agents and tool use → General LLM agent frameworks and benchmarks
PhysAgent: Automating Physics-Based 4D Synthesis via Trajectory-Grounded Multi-Agent FeedbackLLM agents and tool use → Multimodal and embodied agents
BLUE: Toward Better Language Use in Efficient Vision-Language-Action Models for Autonomous DrivingVision-language-action models → VLA for autonomous driving +1
Activation Steering Induces Emergent Misalignment: A More Comprehensive EvaluationMechanistic interpretability → Activation steering methods
SkillHone: A Harness for Continual Agent Skill Evolution Through Persistent Decision HistoryLLM agents and tool use → Agent skill learning and evolution
WaveDiT: Distribution-Aware Wavelet Flow Matching for Efficient 3D Brain MRI SynthesisDiffusion models → Medical imaging and segmentation
Data Agents Under Attack: Vulnerabilities in LLM-Driven Analytical SystemsLLM agents and tool use → Agent threats, guardrails and governance
Latent Diffusion Policy: Shaping Latent Spaces for Diffusion-Based Robotic ManipulationDiffusion models → Diffusion and flow policies for robotics
From Player to Master: Enhancing Test-Time Learning of LLM Agents via Reinforcement Learning over MemoryLLM agents and tool use → Agent memory systems
FiberTune: Preserving Action-Fiber Visual Residuals in Vision-Language-Action Fine-TuningVision-language-action models → VLA architectures and generalization
A retrieval conditioned rebinding circuit for dynamic entity tracking in large language modelsMechanistic interpretability → Activation steering methods
Learnable Token Sparsification for Efficient Gigapixel Whole Slide Image ReasoningLLM reasoning → Multimodal reasoning +1
SpectrumKV: Per-Token Mixed-Precision KV Cache Transfer for Prefill-Decode Disaggregated LLM ServingEfficient LLM inference and serving → KV cache optimization
Towards Long-Horizon Vessel Trajectory and Destination Forecasting with Reasoning Large Language ModelsLLM reasoning → Reinforcement learning for reasoning
Harnessing Streaming Video in the WildMultimodal LLMs / vision-language models → Video understanding with MLLMs
HARBOR: A Harness Framework for Agentic Robot Reinforcement LearningLLM agents and tool use → Multimodal and embodied agents
Reinforcement Learning for Flow-Matching Policies with Density TransportDiffusion models → Diffusion and flow policies for robotics +1
Distilling LLM Reasoning into an Interpretable Policy Tree for Human-AI CollaborationLLM reasoning → Reasoning for agents, planning and code +1
Calibration of Structured Ignorance Certificates for Diagnosing Unknown Unknowns in Reasoning ModelsLLM reasoning → Test-time compute and efficient thinking
FAWAM: Force-Aware World Action Models for Closed-Loop Contact-Rich ManipulationWorld models → Embodied and robot world models
PAEC: Position-Aware Entropy Calibration for LLM Reasoning in RLVRLLM reasoning → Reinforcement learning for reasoning
AgentTrust: A Self-Improving Trust Layer for AI-Agent ActionsLLM agents and tool use → Agent threats, guardrails and governance +1
Autonomous Aerial Manipulation via Contextual Contrastive Meta Reinforcement LearningDeep reinforcement learning methods → RL for robot navigation and UAV control
ForesightSafety-SAGE:A Fully Automated Scenario Generation and Safety Evaluation Framework for LLM AgentsLLM agents and tool use → Agent threats, guardrails and governance
GEAR-VLA: Learning Geometry-Aware Action Representations for Generalizable Robotic ManipulationVision-language-action models → 3D and spatial-aware VLA
DriveReward: A Comprehensive Dataset and Generative Vision-Language Reward Model for Autonomous DrivingEnd-to-end autonomous driving → Reasoning VLA and VLM driving policies +1
Two Bridges, One Pathway: From VLMs to Generalizable VLAs with Embodied Trajectory-Coupled DataMultimodal LLMs / vision-language models → Embodied and 3D vision-language models
Towards End to End Motion Planning and Execution for Autonomous Underwater Vehicles Using Reinforcement LearningDeep reinforcement learning methods → RL for robot navigation and UAV control
Look Less, Reason More: Block-wise Attention Skipping for Efficient Multimodal LLMsLLM reasoning → Multimodal reasoning +1
Back on Track: Aligning Rewards and States for Reasoning in Diffusion Large Language ModelsLLM reasoning → Diffusion reasoning and test-time scaling
Projecting the Emerging Mindset of SWE Agent by Launching a Wild Code Understanding JourneyLLM agents and tool use → Software engineering coding agents
SAEExplainer: Interpreting SAE Features with Activation-Guided Preference OptimizationLLM alignment / RLHF / preference optimisation → Reward modeling and RLHF +1
Adaptive Loss Balancing for Noise-Robust GRPO in Generative RecommendationLLM reasoning → Reinforcement learning for reasoning
More Yap Less Meaning: Uncovering Self-Improvement Behavior in SLMsSynthetic data and self-improvement for LLMs → Synthetic reasoning data and bootstrapping
LUNA-AD: Lightweight Uncertainty-Aware Language Model with Lifelong Learning for Autonomous DrivingLLM reasoning → Chain-of-thought reasoning for driving +1
Nonparametric undirected graphical model selection using diffusion modelsDiffusion models → Flow matching, time series and science
TVI-CoT: Text-Visual Interleaved Chain-of-Thought Reasoning for Multimodal UnderstandingLLM reasoning → Multimodal reasoning
Personalized and Robust Proactive Robot Assistance with Uncertainty-Guided LLM ReasoningLLM reasoning → Spatial and embodied reasoning
The Consistency Illusion: How Multi-Agent Debate Hides Reasoning MisalignmentLLM reasoning → Clinical and biomedical applications +1
Beyond Linear Activation Steering: Invertible Latent Transformations for Controlling LLM BehaviorMechanistic interpretability → Activation steering methods
Sycophancy as a Multilingual Alignment Failure: How Safety Degrades Across Languages, Topics, and ModelsLLM alignment / RLHF / preference optimisation → Safety alignment and jailbreak attacks
Sparrow: Sparse Rollout for Stable and Efficient Long-context RL of Large Language ModelsLLM reasoning → Reinforcement learning for reasoning
Improving Bayesian Optimization via Training-Aware Conditional Diffusion ModelsDiffusion models → Diffusion foundations and sampling
CACR:Reinforcing Temporal Answer Grounding in Instructional Video via Candidate-Aware Causal ReasoningLLM reasoning → Multimodal reasoning
Trajectory-Refined DistillationKnowledge distillation → On-policy and self-distillation for LLMs
TinyGiantALM: A Compact Audio-Language Model for Intent-Aware Reasoning under Resource ConstraintsLLM reasoning → Audio-language model reasoning
CheXanatomy: Anatomy-Aware Vision-Language Modeling for Chest RadiographsMultimodal LLMs / vision-language models → Medical vision-language models
PACT: Self-Evolving Physical Safety Alignment for Diffusion Policies in Embodied ManipulationDiffusion models → Diffusion and flow policies for robotics +1
AsyncLane: Decoupling Refinement from Advancement in Diffusion Language Model DecodingDiffusion models → Discrete and text diffusion
TimpaTeks: Automatic In-place Text Sequence Modification via Diffusion Language Model SteeringDiffusion models → Discrete and text diffusion +1
When Correct Decisions Hide Internal Stress: Decision-State Probing in Multimodal Language ModelsMultimodal LLMs / vision-language models → MLLM training and visual instruction tuning
STAR-KV: Low-Rank KV Cache Compression via Soft Thresholding for Adaptive Rank ControlEfficient LLM inference and serving → KV cache optimization
TT-DAC-PS: Twin-Target Deterministic Actor-Critic with Policy Smoothing for Optimal Trade ExecutionDeep reinforcement learning methods → Adversarial robustness and RL in finance
NE-BERT: A Multilingual Language Model for Nine Northeast Indian Languages
Abliteration Mitigation via Refusal Aliases
Crystalis: Progressive Nucleation and Semantic Annealing for Coordinated Multi-View Visualization Generation
Platooning Connected, Autonomous, and Human-Driven Vehicles: A Deep Reinforcement Learning-based Approach
IOAH3: Importance-Driven Adaptive Spatial Partitioning
Where Does Texture Evidence Live in SAM? Features, Proposal Masks, and Texture Segmentation
Sub-Semantic Image Segmentation
Beyond Self-Attention: Sub-Quadratic Vision Transformers for Fast Image Captioning
nCMD: Benign-Anchored Feature Selection for Imbalanced Network Intrusion Detection

All 249