Papers

New on arXiv

Mon 8 Dec 2025

349 papers

Retrieval-Augmented Generation Assistant for Anatomical Pathology LaboratoriesRetrieval-augmented generation → Medical and biomedical RAG
Response-Based Knowledge Distillation for Multilingual Jailbreak Prevention Unwittingly Compromises SafetyLLM alignment / RLHF / preference optimisation → Safety alignment and jailbreak attacks +2
Reflection-Driven Self-Optimization 6G Agentic AI RAN via Simulation-in-the-Loop WorkflowsLLM agents and tool use → Agentic AI for wireless networks
Inferring Latent Market Forces: Evaluating LLM Detection of Gamma Exposure Patterns via Obfuscation TestingLLM reasoning → Reasoning for agents, planning and code
A Bayesian latent class reinforcement learning framework to capture adaptive, feedback-driven travel behaviourDeep reinforcement learning methods → RL for autonomous driving and traffic
SepsisSuite: Beyond Risk Stratification -- A Comparative Analysis of Deep Fusion vs. Expert Stacking for Prescriptive Sepsis AIMixture-of-experts → Medical and biomedical MoE
Attention as Binding: A Vector-Symbolic Perspective on Transformer ReasoningLLM reasoning → Reasoning benchmarks and evaluation
Meta Hierarchical Reinforcement Learning for Scalable Resource Management in O-RANDeep reinforcement learning methods → RL for energy, networks and operations
PoultryTalk: A Multi-modal Retrieval-Augmented Generation (RAG) System for Intelligent Poultry Management and Decision SupportRetrieval-augmented generation → Time series and agricultural RAG
Deterministic World Models for Closed-loop Reachability Analysis of End-to-End Vision-based ControlWorld models → Embodied and robot world models
Scalable Offline Model-Based RL with Action ChunksOffline RL and imitation learning from data → Offline RL algorithms
AgentCrypt: Advancing Privacy and (Secure) Computation in AI Agent CollaborationLLM agents and tool use → Agent threats, guardrails and governance
Unveiling Latent Knowledge in Chemistry Language Models through Sparse AutoencodersMechanistic interpretability → Interpreting biomedical and science models
An Introduction to Deep Reinforcement and Imitation LearningOffline RL and imitation learning from data → Imitation learning methods +1
CLARITY: Medical World Model for Guiding Treatment Decisions by Simulating Context-Aware Disease TrajectoriesWorld models → Medical and surgical world models
Toward an AI Reasoning-Enabled System for Patient-Clinical Trial MatchingLLM reasoning → Clinical and biomedical applications
FRIEDA: Benchmarking Multi-Step Cartographic Reasoning in Vision-Language ModelsLLM reasoning → Spatial and embodied reasoning +1
Benchmarking Offline Multi-Objective Reinforcement Learning in Critical CareOffline RL and imitation learning from data → Safe offline RL
SkipKV: Selective Skipping of KV Generation and Storage for Efficient Inference with Large Reasoning ModelsLLM reasoning → Test-time compute and efficient thinking +1
VLD: Visual Language Goal Distance for Reinforcement Learning NavigationDeep reinforcement learning methods → RL for robot navigation and UAV control
DeepCode: Open Agentic CodingLLM agents and tool use → Software engineering coding agents
CFD-copilot: leveraging domain-adapted large language model and model context protocol to enhance simulation automationLLM agents and tool use → Web, search, and research agents
One Layer Is Enough: Adapting Pretrained Visual Encoders for Image GenerationDiffusion models → Diffusion foundations and sampling
Auditing Games for SandbaggingLLM safety and security → Prompt injection and agent security
Collaborative Causal Sensemaking: Closing the Complementarity Gap in Human-AI Decision SupportLLM agents and tool use → General LLM agent frameworks and benchmarks
ReasonBENCH: Benchmarking the (In)Stability of LLM ReasoningLLM reasoning → Test-time compute and efficient thinking
Automating High Energy Physics Data Analysis with LLM-Powered AgentsLLM agents and tool use → Web, search, and research agents
On the Interplay of Pre-Training, Mid-Training, and RL on Reasoning Language ModelsLLM reasoning → Reinforcement learning for reasoning
TROJail: Trajectory-Level Optimization for Multi-Turn Large Language Model Jailbreaks with Process RewardsLLM safety and security → Jailbreaking
UltrasODM: A Dual Stream Optical Flow Mamba Network for 3D Freehand Ultrasound ReconstructionState-space sequence models (S4, Mamba) → Mamba for medical image analysis
DiffusionDriveV2: Reinforcement Learning-Constrained Truncated Diffusion Modeling in End-to-End Autonomous DrivingDiffusion models → Diffusion and flow policies for robotics +2
SpatialDreamer: Incentivizing Spatial Reasoning via Active Mental ImageryLLM reasoning → Spatial and embodied reasoning +1
SAVE: Sparse Autoencoder-Driven Visual Information Enhancement for Mitigating Object HallucinationMultimodal LLMs / vision-language models → Hallucination mitigation in LVLMs +1
Each Prompt Matters: Scaling Reinforcement Learning Without Wasting Rollouts on Hundred-Billion-Scale MoELLM reasoning → Reinforcement learning for reasoning +2
Delay-Aware Diffusion Policy: Bridging the Observation-Execution Gap in Dynamic TasksDiffusion models → Diffusion and flow policies for robotics
HalluShift++: Bridging Language and Vision through Internal Representation Shifts for Hierarchical Hallucinations in MLLMsMultimodal LLMs / vision-language models → Hallucination mitigation in LVLMs
Depth-Wise Activation Steering for Honest Language ModelsMechanistic interpretability → Probing deception, bias and beliefs
Reliable agent engineering should integrate machine-compatible organizational principlesLLM agents and tool use → General LLM agent frameworks and benchmarks
Optimization-Guided Diffusion for Interactive Scene GenerationDiffusion models → Diffusion and flow policies for robotics
Exploring Test-time Scaling via Prediction Merging on Large-Scale RecommendationLLM reasoning → Test-time compute and efficient thinking
The Agent Capability Problem: Predicting Solvability Through Information-Theoretic BoundsLLM agents and tool use → General LLM agent frameworks and benchmarks
Comparative Analysis and Parametric Tuning of PPO, GRPO, and DAPO for LLM Reasoning EnhancementLLM reasoning → Reinforcement learning for reasoning
See Once, Then Act: Vision-Language-Action Model with Task Learning from One-Shot Video DemonstrationsVision-language-action models → Human-video and latent-action pretraining
When Token Pruning is Worse than Random: Understanding Visual Token Information in VLLMsEfficient LLM inference and serving → Visual token pruning in multimodal models
Toward More Reliable Artificial Intelligence: Reducing Hallucinations in Vision-Language ModelsMultimodal LLMs / vision-language models → Hallucination mitigation in LVLMs
ReLaX: Reasoning with Latent Exploration for Large Reasoning ModelsLLM reasoning → Test-time compute and efficient thinking
VulnLLM-R: Specialized Reasoning LLM with Agent Scaffold for Vulnerability DetectionLLM reasoning → Reasoning for agents, planning and code +1
Model-Based Reinforcement Learning Under ConfoundingDeep reinforcement learning methods → RL theory and sample complexity
TPA: Next Token Probability Attribution for Detecting Hallucinations in RAGRetrieval-augmented generation → Hallucination detection in RAG
ControlVP: Interactive Geometric Refinement of AI-Generated Images with Consistent Vanishing PointsDiffusion models → 3D, point cloud and scene generation
SJD++: Improved Speculative Jacobi Decoding for Training-free Acceleration of Discrete Auto-regressive Text-to-Image GenerationEfficient LLM inference and serving → Fast autoregressive visual generation
MultiMotion: Multi Subject Video Motion Transfer via Video Diffusion TransformerDiffusion models → Video generation
How Do LLMs Fail In Agentic Scenarios? A Qualitative Analysis of Success and Failure Scenarios of Various LLMs in Agentic SimulationsLLM agents and tool use → General LLM agent frameworks and benchmarks
Single-step Diffusion-based Video Coding with Semantic-Temporal GuidanceDiffusion models → Diffusion for compression and wireless
Enhancing Agentic RL with Progressive Reward Shaping and Value-based Sampling Policy OptimizationLLM reasoning → Reasoning for agents, planning and code +1
Living the Novel: A System for Generating Self-Training Timeline-Aware Conversational Agents from NovelsLLM agents and tool use → Multimodal and embodied agents +1
Affordance Field Intervention: Enabling VLAs to Escape Memory Traps in Robotic ManipulationVision-language-action models → Reasoning and chain-of-thought in VLA
VideoCoF: Unified Video Editing with Temporal ReasonerLLM reasoning → Multimodal reasoning
Gait-Adaptive Perceptive Humanoid Locomotion with Real-Time Under-Base Terrain ReconstructionDeep reinforcement learning methods → RL for robot locomotion and manipulation
Understanding LLM Agent Behaviours via Game Theory: Strategy Recognition, Biases and Multi-Agent DynamicsLLM agents and tool use → Social simulation and multi-agent debate
Native Parallel Reasoner: Reasoning in Parallelism via Self-Distilled Reinforcement LearningLLM reasoning → Test-time compute and efficient thinking
KAN-Dreamer: Benchmarking Kolmogorov-Arnold Networks as Function Approximators in World ModelsWorld models → Latent dynamics models and model-based RL +1
LocalSearchBench: Benchmarking Agentic Search in Real-World Local Life ServicesLLM reasoning → Knowledge- and tool-augmented reasoning +1
Training Language Models to Use Prolog as a ToolLLM reasoning → Reinforcement learning for reasoning
Control and Reinforcement Learning through the Lens of Optimization: An Algorithmic PerspectiveDeep reinforcement learning methods → RL theory and sample complexity
LUNE: Efficient LLM Unlearning via LoRA Fine-Tuning with Negative ExamplesLLM safety and security → Machine unlearning
Recover-to-Forget: Gradient Reconstruction from LoRA for Efficient LLM UnlearningLLM safety and security → Machine unlearning
ESPADA: Execution Speedup via Semantics Aware Demonstration Data Downsampling for Imitation LearningOffline RL and imitation learning from data → Robot manipulation imitation learning
Structure-Aware Feature Rectification with Region Adjacency Graphs for Training-Free Open-Vocabulary Semantic SegmentationMultimodal LLMs / vision-language models → Prompt learning and adaptation of CLIP
A Geometric Unification of Concept Learning with Concept ConesMechanistic interpretability → Sparse autoencoder training and evaluation
Debiasing Diffusion Priors via 3D Attention for Consistent Gaussian SplattingDiffusion models → Image generation, text-to-image and editing
Venus: An Efficient Edge Memory-and-Retrieval System for VLM-based Online Video UnderstandingMultimodal LLMs / vision-language models → Video understanding with MLLMs
PrivORL: Differentially Private Synthetic Dataset for Offline Reinforcement LearningDiffusion models → Diffusion and flow policies for robotics +1
Towards Accurate UAV Image Perception: Guiding Vision-Language Models with Stronger Task PromptsMultimodal LLMs / vision-language models → Embodied and 3D vision-language models
Equivariant Diffusion for Crystal Structure PredictionDiffusion models → Flow matching, time series and science
Geo3DVQA: Evaluating Vision-Language Models for 3D Geospatial Reasoning from Aerial ImageryLLM reasoning → Spatial and embodied reasoning +1
RVLF: A Reinforcing Vision-Language Framework for Gloss-Free Sign Language TranslationLLM reasoning → Multimodal reasoning
SINRL: Socially Integrated Navigation with Reinforcement Learning using Spiking Neural NetworksDeep reinforcement learning methods → RL for robot navigation and UAV control
Unified Camera Positional Encoding for Controlled Video GenerationWorld models → Video-generation world models and simulators
Dropout Prompt Learning: Towards Robust and Adaptive Vision-Language ModelsMultimodal LLMs / vision-language models → Prompt learning and adaptation of CLIP
ReLKD: Inter-Class Relation Learning with Knowledge Distillation for Generalized Category DiscoveryKnowledge distillation → General knowledge distillation methods
Unsupervised Single-Channel Audio Separation with Diffusion Source PriorsDiffusion models → Audio, speech and music
Pay Less Attention to Function Words for Free Robustness of Vision-Language ModelsMultimodal LLMs / vision-language models → MLLM jailbreaks and adversarial robustness
NeSTR: A Neuro-Symbolic Abductive Framework for Temporal Reasoning in Large Language ModelsLLM reasoning → Knowledge- and tool-augmented reasoning
VFM-VLM: Vision Foundation Model and Vision Language Model based Visual Comparison for 3D Pose EstimationMultimodal LLMs / vision-language models → Embodied and 3D vision-language models
Sample from What You See: Visuomotor Policy Learning via Diffusion Bridge with Observation-Embedded Stochastic Differential EquationDiffusion models → Diffusion and flow policies for robotics +1
MMRPT: MultiModal Reinforcement Pre-Training via Masked Vision-Dependent ReasoningLLM reasoning → Multimodal reasoning
Understanding Diffusion Models via Code ExecutionDiffusion models → Diffusion foundations and sampling
Generating Storytelling Images with Rich Chains-of-ReasoningLLM reasoning → Multimodal reasoning
Multilingual Agent-Based World Modeling for Social ScienceWorld models → World models for language agents
START: Spatial and Textual Learning for Chart UnderstandingMultimodal LLMs / vision-language models → Chart understanding and anomaly detection
Using Vision-Language Models as Proxies for Social Intelligence in Human-Robot InteractionMultimodal LLMs / vision-language models → Embodied and 3D vision-language models
SPACE: Noise Contrastive Estimation Stabilizes Self-Play Fine-Tuning for Large Language ModelsSynthetic data and self-improvement for LLMs → Self-rewarding alignment and self-training
Towards Unified Semantic and Controllable Image Fusion: A Diffusion Transformer ApproachDiffusion models → Image generation, text-to-image and editing
CHIMERA: Adaptive Cache Injection and Semantic Anchor Prompting for Zero-shot Image Morphing with Morphing-oriented MetricsDiffusion models → Image generation, text-to-image and editing
Think-Reflect-Revise: A Policy-Guided Reflective Framework for Safety Alignment in Large Vision Language ModelsLLM reasoning → Reasoning for agents, planning and code +3
A Large-Scale Multimodal Dataset and Benchmarks for Human Activity Scene Understanding and ReasoningLLM reasoning → Multimodal reasoning
TrajMoE: Scene-Adaptive Trajectory Planning with Mixture of Experts and Reinforcement LearningMixture-of-experts → MoE in reinforcement learning and robotics +1
DART: Leveraging Multi-Agent Disagreement for Tool Recruitment in Multimodal ReasoningLLM reasoning → Multimodal reasoning +2
Mimir: Hierarchical Goal-Driven Diffusion with Uncertainty Propagation for End-to-End Autonomous DrivingEnd-to-end autonomous driving → Imitation-learning end-to-end planners

All 349