Papers

New on arXiv

Tue 24 Feb 2026

411 papers

Mirror Descent-Type Algorithms for the Variational Inequality Problem with Functional ConstraintsDeep reinforcement learning methods → RL theory and sample complexity
Caesar: Deep Agentic Web Exploration for Creative Answer SynthesisLLM agents and tool use → Web, search, and research agents
ERA: Evidence-based Reliability Alignment for Honest Retrieval-Augmented GenerationRetrieval-augmented generation → RAG robustness and adaptive retrieval
How Robustly do LLMs Understand Execution Semantics?World models → World models for language agents +1
Turing Test on Screen: A Benchmark for Mobile GUI Agent HumanizationLLM agents and tool use → GUI and mobile agents
MineDraft: A Framework for Batch Parallel Speculative DecodingEfficient LLM inference and serving → Speculative decoding
Beyond Accuracy: An Explainability-Driven Analysis of Harmful Content DetectionLLM safety and security → Safety alignment, refusal and guardrails
DynaRAG: Bridging Static and Dynamic Knowledge in Retrieval-Augmented GenerationRetrieval-augmented generation → RAG robustness and adaptive retrieval
NeuroNarrator: A Generalist EEG-to-Text Foundation Model for Clinical Interpretation via Spectro-Spatial Grounding and Temporal State-Space ReasoningLLM reasoning → Clinical and biomedical applications
Physics-Informed Video Diffusion For Shallow Water EquationsDiffusion models → Flow matching, time series and science
Distilling Deep Reinforcement Learning into Interpretable Fuzzy Rules: An Explainable AI FrameworkDeep reinforcement learning methods → Interpretable and logic-guided RL
Training-Free Agentic AI: Probabilistic Control and Coordination in Multi-Agent LLM SystemsLLM reasoning → Reasoning for agents, planning and code +1
The Ghost in the Grammar: Methodological Anthropomorphism in AI Safety EvaluationsLLM agents and tool use → General LLM agent frameworks and benchmarks
Steering at the Source: Style Modulation Heads for Robust Persona ControlMechanistic interpretability → Activation steering methods
Graph Property Inference in Small Language Models: Effects of Representation and Reasoning StrategyLLM reasoning → Knowledge- and tool-augmented reasoning
Mechanistically Guided LoRA Improves Paraphrase Consistency in Medical Vision-Language ModelsMultimodal LLMs / vision-language models → Medical vision-language models
Disentangled Hierarchical VAE for 3D Human-Human Interaction GenerationDiffusion models → Video generation
Evaluating Theory of Mind and Internal Beliefs in LLM-Based Multi-Agent SystemsLLM agents and tool use → General LLM agent frameworks and benchmarks
From Scale to Speed: Adaptive Test-Time Scaling for Image EditingLLM reasoning → Multimodal reasoning
Steering Away from Memorization: Reachability-Constrained Reinforcement Learning for Text-to-Image DiffusionDiffusion models → Diffusion and flow policies for robotics +1
TinyVLM: Zero-Shot Object Detection on Microcontrollers via Vision-Language Distillation with Matryoshka EmbeddingsMultimodal LLMs / vision-language models → Prompt learning and adaptation of CLIP
Detoxifying LLMs via Representation Erasure-Based Preference OptimizationLLM alignment / RLHF / preference optimisation → Safety alignment and jailbreak attacks
Self-Purification Mitigates Backdoors in Multimodal Diffusion Language ModelsDiffusion models → Robustness, privacy and adversarial aspects +1
Analysis of LLMs Against Prompt Injection and Jailbreak AttacksLLM safety and security → Jailbreaking
MERRY: Semantically Decoupled Evaluation of Multimodal Emotional and Role Consistencies of Role-Playing AgentsLLM agents and tool use → Social simulation and multi-agent debate
Adversarial Intent is a Latent Variable: Stateful Trust Inference for Securing Multimodal Agentic RAGLLM agents and tool use → Agent threats, guardrails and governance
VLA Knows Its Limits: Adaptive Execution Horizons for Robot PoliciesVision-language-action models → Efficient VLA inference and adaptation
MINAR: Mechanistic Interpretability for Neural Algorithmic ReasoningLLM reasoning → Reasoning benchmarks and evaluation +1
Causal Decoding for Hallucination-Resistant Multimodal Large Language ModelsMultimodal LLMs / vision-language models → Hallucination mitigation in LVLMs
Synergizing Understanding and Generation with Interleaved Analyzing-Drafting ThinkingLLM reasoning → Multimodal reasoning +1
Provably Safe Generative Sampling with Constricting Barrier FunctionsDiffusion models → Diffusion foundations and sampling
PSF-Med: Measuring and Explaining Paraphrase Sensitivity in Medical Vision Language ModelsMultimodal LLMs / vision-language models → Medical vision-language models
On the Structural Non-Preservation of Epistemic Behaviour under Policy TransformationDeep reinforcement learning methods → Deep RL algorithms and representations
Overconfident Errors Need Stronger Correction: Asymmetric Confidence Penalties for Reinforcement LearningLLM reasoning → Reinforcement learning for reasoning +1
Exploring Vision-Language Models for Open-Vocabulary Zero-Shot Action SegmentationMultimodal LLMs / vision-language models → Video understanding with MLLMs
MMLoP: Multi-Modal Low-Rank Prompting for Efficient Vision-Language AdaptationMultimodal LLMs / vision-language models → Prompt learning and adaptation of CLIP
Momentum Memory for Knowledge Distillation in Computational PathologyKnowledge distillation → Distillation for medical imaging
MemoPhishAgent: Memory-Augmented Multi-Modal LLM Agent for Phishing URL DetectionLLM agents and tool use → Agent threats, guardrails and governance
Black-Box Reliability Certification for AI Agents via Self-Consistency Sampling and Conformal CalibrationLLM reasoning → Test-time compute and efficient thinking +1
Towards Controllable Video Synthesis of Routine and Rare OR EventsDiffusion models → Medical imaging and segmentation
A Hierarchical Multi-Agent System for Autonomous Discovery in Geoscientific Data ArchivesLLM agents and tool use → Web, search, and research agents
Alignment-Weighted DPO: A principled reasoning approach to improve safety alignmentLLM reasoning → Reasoning for agents, planning and code +1
HorizonForge: Driving Scene Editing with Any Trajectories and Any VehiclesDiffusion models → Diffusion and flow policies for robotics
Tool-R0: Self-Evolving LLM Agents for Tool-Learning from Zero DataLLM agents and tool use → Reinforcement learning for LLM agents +1
Uncertainty-Aware Diffusion Model for Multimodal Highway Trajectory Prediction via DDIM SamplingDiffusion models → Diffusion and flow policies for robotics
Shared Nature, Unique Nurture: PRISM for Pluralistic Reasoning via In-context Structure ModelingLLM reasoning → Test-time compute and efficient thinking
ToolMATH: A Diagnostic Benchmark for Long-Horizon Tool Use under Systematic Tool-Catalog ConstraintsLLM agents and tool use → Tool learning and function calling
Squint: Fast Visual Reinforcement Learning for Sim-to-Real RoboticsDeep reinforcement learning methods → RL for robot locomotion and manipulation
Why Pass@k Optimization Can Degrade Pass@1: Prompt Interference in LLM Post-trainingLLM reasoning → Reinforcement learning for reasoning
Spa3R: Predictive Spatial Field Modeling for 3D Visual ReasoningLLM reasoning → Spatial and embodied reasoning +1
The Diffusion Duality, Chapter II: Psi-SamplersDiffusion models → Discrete and text diffusion
Memory Undone: Between Knowing and Not Knowing in Data SystemsLLM safety and security → Machine unlearning
NoRD: A Data-Efficient Vision-Language-Action Model that Drives without ReasoningLLM reasoning → Chain-of-thought reasoning for driving +1
ActionReasoning: Robot Action Reasoning in 3D Space with LLM for Robotic Brick StackingLLM reasoning → Spatial and embodied reasoning
SELAUR: Self Evolving LLM Agent via Uncertainty-aware RewardsLLM agents and tool use → Reinforcement learning for LLM agents
HALO: A Unified Vision-Language-Action Model for Embodied Multimodal Chain-of-Thought ReasoningLLM reasoning → Spatial and embodied reasoning +1
Scaling State-Space Models on Multiple GPUs with Tensor ParallelismState-space sequence models (S4, Mamba) → Core SSM architectures for language
LUMEN: Longitudinal Multi-Modal Radiology Model for Prognosis and DiagnosisMultimodal LLMs / vision-language models → Medical vision-language models
UDVideoQA: A Traffic Video Question Answering Dataset for Multi-Object Spatio-Temporal Reasoning in Urban DynamicsLLM reasoning → Chain-of-thought reasoning for driving
"Are You Sure?": An Empirical Study of Human Perception Vulnerability in LLM-Driven Agentic SystemsLLM agents and tool use → Agent threats, guardrails and governance
Cooperative-Competitive Team Play of Real-World Craft RobotsDeep reinforcement learning methods → RL for robot locomotion and manipulation
Prompt-Level Distillation: A Non-Parametric Alternative to Model Fine-Tuning for Efficient ReasoningLLM reasoning → Prompting and chain-of-thought reasoning
Turning Semantics into Topology: LLM-Driven Attribute Augmentation for Collaborative FilteringLLM reasoning → LLM reasoning for recommendation
Localized Dynamics-Aware Domain Adaption for Off-Dynamics Offline Reinforcement LearningOffline RL and imitation learning from data → Offline RL algorithms
Tool Use Reduces Depth-Induced Collapse in OOD ReasoningLLM reasoning → Test-time compute and efficient thinking
VAUQ: Vision-Aware Uncertainty Quantification for LVLM Self-EvaluationMultimodal LLMs / vision-language models → Hallucination mitigation in LVLMs
OCR-Agent: Agentic OCR with Capability and Memory ReflectionMultimodal LLMs / vision-language models → Document understanding and multimodal RAG +1
LogicGraph : Benchmarking Multi-Path Logical Reasoning via Neuro-Symbolic Generation and VerificationLLM reasoning → Reasoning benchmarks and evaluation
Not Just What's There: Enabling CLIP to Comprehend Negated Visual Descriptions Without Fine-tuningMultimodal LLMs / vision-language models → Prompt learning and adaptation of CLIP
Matching Multiple Experts: On the Exploitability of Multi-Agent Imitation LearningOffline RL and imitation learning from data → Imitation learning methods
From Perception to Action: An Interactive Benchmark for Vision ReasoningLLM reasoning → Spatial and embodied reasoning
Notes-to-Self: Scratchpad Augmented VLAs for Memory Dependent Manipulation TasksVision-language-action models → Reasoning and chain-of-thought in VLA
VII: Visual Instruction Injection for Jailbreaking Image-to-Video Generation ModelsLLM safety and security → Vision-language and text-to-image safety
CrystaL: Spontaneous Emergence of Visual Latents in MLLMsMultimodal LLMs / vision-language models → VLM benchmarks and visual reasoning
Toward an Agentic Infused Software EcosystemLLM agents and tool use → Software engineering coding agents
Linear Reasoning vs. Proof by Cases: Obstacles for Large Language Models in FOL Problem SolvingLLM reasoning → Mathematical reasoning
Are Multimodal Large Language Models Good Annotators for Image Tagging?Multimodal LLMs / vision-language models → MLLM training and visual instruction tuning
See and Fix the Flaws: Enabling VLMs and Diffusion Models to Comprehend Visual Artifacts via Agentic Data SynthesisDiffusion models → Image generation, text-to-image and editing +1
The Art of Efficient Reasoning: Data, Reward, and OptimizationLLM reasoning → Test-time compute and efficient thinking
UFO: Unifying Feed-Forward and Optimization-based Methods for Large Driving Scene ModelingEnd-to-end autonomous driving → Generative simulation for driving
Architecting AgentOS: From Token-Level Context to Emergent System-Level IntelligenceLLM reasoning → Reasoning for agents, planning and code
HELP: HyperNode Expansion and Logical Path-Guided Evidence Localization for Accurate and Efficient GraphRAGRetrieval-augmented generation → Knowledge-graph and structured-knowledge RAG
Airavat: An Agentic Framework for Internet MeasurementLLM agents and tool use → Software engineering coding agents
ParkDiffusion++: Ego Intention Conditioned Joint Multi-Agent Trajectory Prediction for Automated Parking using Diffusion ModelsDiffusion models → Diffusion and flow policies for robotics
Task-oriented grasping for dexterous robots using postural synergies and reinforcement learningDeep reinforcement learning methods → RL for robot locomotion and manipulation
Transcoder Adapters for Reasoning-Model DiffingLLM reasoning → Test-time compute and efficient thinking +1
SpatiaLQA: A Benchmark for Evaluating Spatial Logical Reasoning in Vision-Language ModelsLLM reasoning → Spatial and embodied reasoning +1
When Safety Collides: Resolving Multi-Category Harmful Conflicts in Text-to-Image Diffusion via Adaptive Safety GuidanceDiffusion models → Robustness, privacy and adversarial aspects +1
Diagnosing Causal Reasoning in Vision-Language Models via Structured Relevance GraphsLLM reasoning → Multimodal reasoning +1
SoK: Agentic Skills -- Beyond Tool Use in LLM AgentsLLM agents and tool use → Agent skill learning and evolution
On the Explainability of Vision-Language Models in Art HistoryMultimodal LLMs / vision-language models → VLM benchmarks and visual reasoning
Training-Free Multi-Concept Image EditingDiffusion models → Image generation, text-to-image and editing
Regret-Guided Search Control for Efficient Learning in AlphaZeroDeep reinforcement learning methods → Deep RL algorithms and representations
Unseen-Codebases-Domain Data Synthesis and Training Based on Code GraphsSynthetic data and self-improvement for LLMs → Code and instruction data synthesis
VGGDrive: Empowering Vision-Language Models with Cross-View Geometric Grounding for Autonomous DrivingMultimodal LLMs / vision-language models → Driving VLMs and open-vocabulary detection +1
Pipeline for Verifying LLM-Generated Mathematical SolutionsLLM reasoning → Mathematical reasoning
PyVision-RL: Forging Open Agentic Vision Models via RLLLM agents and tool use → Multimodal and embodied agents
RMIT-ADM+S at the MMU-RAG NeurIPS 2025 CompetitionRetrieval-augmented generation → RAG pipeline design and evaluation
CHESS: Context-aware Hierarchical Efficient Semantic Selection for Long-Context LLM InferenceEfficient LLM inference and serving → KV cache optimization
Communication-Inspired Tokenization for Structured Image RepresentationsDiffusion models → Diffusion foundations and sampling

All 411