Papers

New on arXiv

Mon 30 Jun 2025

363 papers

Fuzzing: Randomness? Reasoning! Efficient Directed Fuzzing via Large Language ModelsLLM reasoning → Reasoning for agents, planning and code
PWD: Prior-Guided and Wavelet-Enhanced Diffusion Model for Limited-Angle CTDiffusion models → Medical imaging and segmentation
How Safe Are AI-Generated Patches? A Large-scale Study on Security Risks in LLM and Agentic Automated Program Repair on SWE-benchLLM agents and tool use → Agent threats, guardrails and governance
Introducing Answered with Evidence -- a framework for evaluating whether LLM responses to biomedical questions are founded in evidenceRetrieval-augmented generation → Medical and biomedical RAG
InvisibleInk: High-Utility and Low-Cost Text Generation with Differential PrivacyRetrieval-augmented generation → Privacy and security of RAG
Reinforcement Learning for Automated Cybersecurity Penetration TestingDeep reinforcement learning methods → Adversarial robustness and RL in finance
RAG-R1: Incentivizing the Search and Reasoning Capabilities of LLMs through Multi-query ParallelismLLM reasoning → Knowledge- and tool-augmented reasoning
Flow-Through Tensors: A Unified Computational Graph Architecture for Multi-Layer Transportation Network OptimizationDeep reinforcement learning methods → RL for autonomous driving and traffic
CS-VLM: Compressed Sensing Attention for Efficient Vision-Language Representation LearningMultimodal LLMs / vision-language models → Visual token pruning and efficiency
Optimizing Conversational Product Recommendation via Reinforcement LearningDeep reinforcement learning methods → RL training of language and agent models
When Digital Twins Meet Large Language Models: Realistic, Interactive, and Editable Simulation for Autonomous DrivingEnd-to-end autonomous driving → Generative simulation for driving
Deep Double Q-learningDeep reinforcement learning methods → Deep RL algorithms and representations
Mechanical Intelligence-Aware Curriculum Reinforcement Learning for Humanoids with Parallel ActuationDeep reinforcement learning methods → RL for robot locomotion and manipulation
Feature Integration Spaces: Joint Training Reveals Dual Encoding in Neural Network RepresentationsMechanistic interpretability → Sparse autoencoder training and evaluation
Control-Optimized Deep Reinforcement Learning for Artificially Intelligent Autonomous SystemsDeep reinforcement learning methods → Deep RL algorithms and representations
Gym4ReaL: A Suite for Benchmarking Real-World Reinforcement LearningDeep reinforcement learning methods → Deep RL algorithms and representations
EfficientXLang: Towards Improving Token Efficiency Through Cross-Lingual ReasoningLLM reasoning → Reasoning benchmarks and evaluation
Linearly Decoding Refused Knowledge in Aligned Language ModelsLLM alignment / RLHF / preference optimisation → Safety alignment and jailbreak attacks +2
Sim2Real Diffusion: Leveraging Foundation Vision Language Models for Adaptive Automated DrivingMultimodal LLMs / vision-language models → Driving VLMs and open-vocabulary detection
Investigating Stochastic Methods for Prosody Modeling in Speech SynthesisDiffusion models → Audio, speech and music
Two-Stage Reasoning-Infused Learning: Improving Classification with LLM-Generated ReasoningLLM reasoning → Audio-language model reasoning
LineRetriever: Planning-Aware Observation Reduction for Web AgentsLLM agents and tool use → Web, search, and research agents
Towards 3D Semantic Image Synthesis for Medical ImagingDiffusion models → Medical imaging and segmentation
Text-to-Level Diffusion Models With Various Text Encoders for Super Mario BrosDiffusion models → Image generation, text-to-image and editing
ChatGPT produces more "lazy" thinkers: Evidence of cognitive engagement declineLLM reasoning → Reasoning benchmarks and evaluation
BlackBoxToBlueprint: Extracting Interpretable Logic from Legacy Systems using Reinforcement Learning and Counterfactual AnalysisDeep reinforcement learning methods → Interpretable and logic-guided RL
Diffusion-Based Image Augmentation for Semantic Segmentation in Outdoor RoboticsDiffusion models → Medical imaging and segmentation
Thinking About Thinking: SAGE-nano's Inverse Reasoning for Self-Aware Language ModelsLLM reasoning → Reasoning benchmarks and evaluation
State and Memory is All You Need for Robust and Reliable AI AgentsLLM agents and tool use → Web, search, and research agents
SPIRAL: Self-Play on Zero-Sum Games Incentivizes Reasoning via Multi-Agent Multi-Turn Reinforcement LearningLLM reasoning → Reasoning for agents, planning and code +2
Epona: Autoregressive Diffusion World Model for Autonomous DrivingWorld models → World models for autonomous driving +2
Navigating with Annealing Guidance Scale in Diffusion SpaceDiffusion models → Diffusion foundations and sampling
DenseWorld-1M: Towards Detailed Dense Grounded Caption in the Real WorldMultimodal LLMs / vision-language models → VLM benchmarks and visual reasoning
Logit-Gap Steering: A Forward-Pass Diagnostic for Alignment RobustnessLLM alignment / RLHF / preference optimisation → Safety alignment and jailbreak attacks
Agent.xpu: Efficient Scheduling of Agentic LLM Workloads on Heterogeneous SoCLLM agents and tool use → Software engineering coding agents
A Survey on Vision-Language-Action Models for Autonomous DrivingMultimodal LLMs / vision-language models → Driving VLMs and open-vocabulary detection +2
Faster Diffusion Models via Higher-Order ApproximationDiffusion models → Diffusion foundations and sampling
Supervised Diffusion-Model-Based PET Image ReconstructionDiffusion models → Medical imaging and segmentation
Constructing Non-Markovian Decision Process via History AggregatorDeep reinforcement learning methods → RL theory and sample complexity
Ella: Embodied Social Agents with Lifelong MemoryLLM agents and tool use → Multimodal and embodied agents
EXPERT: An Explainable Image Captioning Evaluation Metric with Structured ExplanationsMultimodal LLMs / vision-language models → VLM benchmarks and visual reasoning
Provably Efficient and Agile Randomized Q-LearningDeep reinforcement learning methods → RL theory and sample complexity
The Illusion of Progress? A Critical Look at Test-Time Adaptation for Vision-Language ModelsMultimodal LLMs / vision-language models → Prompt learning and adaptation of CLIP
Auto-TA: Towards Scalable Automated Thematic Analysis (TA) via Multi-Agent Large Language Models with Reinforcement LearningLLM agents and tool use → Medical and scientific domain agents
STCLocker: Deadlock Avoidance Testing for Autonomous Driving SystemsEnd-to-end autonomous driving → Adversarial scenarios and safety testing
StreamFlow: Streaming Flow Matching with Block-wise Guided Attention Mask for Speech Token DecodingDiffusion models → Audio, speech and music
StyleDrive: Towards Driving-Style Aware Benchmarking of End-To-End Autonomous DrivingEnd-to-end autonomous driving → Imitation-learning end-to-end planners
LLM Agents Are the Antidote to Walled GardensLLM agents and tool use → General LLM agent frameworks and benchmarks
ADReFT: Adaptive Decision Repair for Safe Autonomous Driving via Reinforcement Fine-TuningEnd-to-end autonomous driving → Reinforcement learning for driving
Bridging the Gap with Retrieval-Augmented Generation: Making Prosthetic Device User Manuals Available in Marginalised LanguagesRetrieval-augmented generation → Medical and biomedical RAG
Unveiling Decision-Making in LLMs for Text Classification : Extraction of influential and interpretable concepts with Sparse AutoencodersMechanistic interpretability → SAE features and their applications
Adapt Your Body: Mitigating Proprioception Shifts in Imitation LearningOffline RL and imitation learning from data → Robot manipulation imitation learning
Graft: Integrating the Domain Knowledge via Efficient Parameter Synergy for MLLMsMultimodal LLMs / vision-language models → MLLM training and visual instruction tuning
Reinforcement Learning for Synchronised Flow Control in a Dual-Gate Resin Infusion SystemDeep reinforcement learning methods → RL with diffusion and flow models
Goal-VLA: Image-Generative VLMs as Object-Centric World Models Empowering Zero-shot Robot ManipulationWorld models → Embodied and robot world models
Thinking with Images for Multimodal Reasoning: Foundations, Methods, and Future FrontiersLLM reasoning → Multimodal reasoning
Beyond Statistical Learning: Exact Learning Is Essential for General IntelligenceLLM reasoning → Reasoning benchmarks and evaluation
Grounding DINO-US-SAM: Text-Prompted Multi-Organ Segmentation in Ultrasound with LoRA-Tuned Vision-Language ModelsMultimodal LLMs / vision-language models → Medical vision-language models
Advancing Multi-Step Mathematical Reasoning in Large Language Models through Multi-Layered Self-Reflection with Auto-PromptingLLM reasoning → Mathematical reasoning
Garbage In, Reasoning Out? Why Benchmark Scores are Unreliable and What to Do About ItLLM reasoning → Reasoning benchmarks and evaluation
VMoBA: Mixture-of-Block Attention for Video Diffusion ModelsDiffusion models → Video generation
A Closer Look at Conditional Prompt Tuning for Vision-Language ModelsMultimodal LLMs / vision-language models → Prompt learning and adaptation of CLIP
Position: Use Sparse Autoencoders to Discover UnknownsMechanistic interpretability → Sparse autoencoder training and evaluation
A Survey on Autonomy-Induced Security Risks in Large Model-Based AgentsLLM agents and tool use → Agent threats, guardrails and governance
Do Thinking Tokens Help or Trap? Towards More Efficient Large Reasoning ModelLLM reasoning → Test-time compute and efficient thinking
Towards the "Digital Me": A vision of authentic Conversational Agents powered by personal Human Digital TwinsLLM agents and tool use → Agent memory systems
Flash-VStream: Efficient Real-Time Understanding for Long Video StreamsMultimodal LLMs / vision-language models → Video understanding with MLLMs
Interpretable Zero-Shot Learning with Locally-Aligned Vision-Language ModelMultimodal LLMs / vision-language models → Prompt learning and adaptation of CLIP
Controllable Reference Guided Diffusion with Local Global Fusion for Real World Remote Sensing Image Super ResolutionDiffusion models → Inverse problems and image restoration
Visual Textualization for Image Prompted Object DetectionMultimodal LLMs / vision-language models → Driving VLMs and open-vocabulary detection
Mamba-FETrack V2: Revisiting State Space Model for Frame-Event based Visual Object TrackingState-space sequence models (S4, Mamba) → Point clouds, detection and remote sensing
Leveraging a Multi-Agent LLM-Based System to Educate Teachers in Hate Incidents ManagementLLM agents and tool use → Educational agents and math problem solving
Multi-Timescale Hierarchical Reinforcement Learning for Unified Behavior and Control of Autonomous DrivingEnd-to-end autonomous driving → Reinforcement learning for driving +1
Radioactive Watermarks in Diffusion and Autoregressive Image Generative ModelsLLM safety and security → Watermarking generated content and models
PAC Bench: Do Foundation Models Understand Prerequisites for Executing Manipulation Policies?Multimodal LLMs / vision-language models → Embodied and 3D vision-language models
DABstep: Data Agent Benchmark for Multi-step ReasoningLLM reasoning → Reasoning for agents, planning and code +1
Subjective Camera 1.0: Bridging Human Cognition and Visual Reconstruction through Sequence-Aware Sketch-Guided DiffusionDiffusion models → 3D, point cloud and scene generation
MDPG: Multi-domain Diffusion Prior Guidance for MRI ReconstructionDiffusion models → Medical imaging and segmentation
SynMotion: Semantic-Visual Adaptation for Motion Customized Video GenerationDiffusion models → Video generation
Pok'eAI: A Goal-Generating, Battle-Optimizing Multi-agent System for Pokemon RedLLM agents and tool use → Reinforcement learning for LLM agents
Learning Modular Exponentiation with TransformersMechanistic interpretability → Circuits and mechanistic analysis
Interactive Reasoning: Visualizing and Controlling Chain-of-Thought Reasoning in Large Language ModelsLLM reasoning → Reasoning benchmarks and evaluation
A Unified Framework for Stealthy Adversarial Generation via Latent Optimization and Transferability EnhancementDiffusion models → Robustness, privacy and adversarial aspects
Efficient Interleaved Speech Modeling through Knowledge DistillationKnowledge distillation → Speech and audio distillation
L0: Reinforcement Learning to Become General AgentsLLM agents and tool use → Reinforcement learning for LLM agents +1
Diffusion Model-based Data Augmentation Method for Fetal Head Ultrasound SegmentationDiffusion models → Medical imaging and segmentation
On the Domain Robustness of Contrastive Vision-Language ModelsMultimodal LLMs / vision-language models → VLM benchmarks and visual reasoning
Act-With-Think: Chunk Auto-Regressive Modeling for Generative RecommendationLLM reasoning → LLM reasoning for recommendation
Unified Multimodal Understanding via Byte-Pair Visual EncodingMultimodal LLMs / vision-language models → MLLM training and visual instruction tuning
Towards Building Private LLMs: Exploring Multi-Node Expert Parallelism on Apple Silicon for Mixture-of-Experts Large Language ModelMixture-of-experts → Efficient MoE training and inference systems
Blending Concepts with Text-to-Image Diffusion ModelsDiffusion models → Image generation, text-to-image and editing
Self-correcting Reward Shaping via Language Models for Reinforcement Learning Agents in GamesLLM agents and tool use → Reinforcement learning for LLM agents
TurboVSR: Fantastic Video Upscalers and Where to Find ThemDiffusion models → Inverse problems and image restoration
SG-LDM: Semantic-Guided LiDAR Generation via Latent-Aligned DiffusionWorld models → World modeling from LiDAR generation +1
CAI: Caption-Sensitive Attention Intervention for Mitigating Object Hallucination in Large Vision-Language ModelsMultimodal LLMs / vision-language models → Hallucination mitigation in LVLMs
Transition Matching: Scalable and Flexible Generative ModelingDiffusion models → Flow matching, time series and science
Evaluating Multi-Agent Defences Against Jailbreaking Attacks on Large Language ModelsLLM agents and tool use → Agent threats, guardrails and governance +1
Metadata, Wavelet, and Time Aware Diffusion Models for Satellite Image Super ResolutionDiffusion models → Inverse problems and image restoration
MMReason: An Open-Ended Multi-Modal Multi-Step Reasoning Benchmark for MLLMs Toward AGILLM reasoning → Multimodal reasoning +1
JAM-Flow: Joint Audio-Motion Synthesis with Flow MatchingDiffusion models → Audio, speech and music

All 363