Papers

New on arXiv

Thu 4 Sep 2025

304 papers

Wavelet Fourier Diffuser: Frequency-Aware Diffusion Model for Reinforcement LearningDiffusion models → Diffusion and flow policies for robotics +1
MEUV: Achieving Fine-Grained Capability Activation in Large Language Models via Mutually Exclusive Unlock VectorsLLM alignment / RLHF / preference optimisation → Safety alignment and jailbreak attacks +1
Mitigating Catastrophic Forgetting and Mode Collapse in Text-to-Image Diffusion via Latent ReplayDiffusion models → Robustness, privacy and adversarial aspects
Resource-Aware Neural Network Pruning Using Graph-based Reinforcement LearningDeep reinforcement learning methods → Deep RL algorithms and representations
3D and 4D World Modeling: A SurveyWorld models → Video-generation world models and simulators
Not All Splits Are Equal: Rethinking Attribute Generalization Across Unrelated CategoriesLLM reasoning → Multimodal reasoning
Visible Yet Unreadable: A Systematic Blind Spot of Vision Language Models Across Writing SystemsMultimodal LLMs / vision-language models → VLM benchmarks and visual reasoning
Murphys Laws of AI Alignment: Why the Gap Always WinsLLM alignment / RLHF / preference optimisation → Reward modeling and RLHF
ThreatGPT: An Agentic AI Framework for Enhancing Public Safety through Threat ModelingLLM agents and tool use → Agent threats, guardrails and governance
Code Like Humans: A Multi-Agent Solution for Medical CodingLLM agents and tool use → Medical and scientific domain agents
Long-Horizon Visual Imitation Learning via Plan and Code ReflectionOffline RL and imitation learning from data → Robot manipulation imitation learning
Between a Rock and a Hard Place: The Tension Between Ethical Reasoning and Safety Alignment in LLMsLLM reasoning → Reasoning for agents, planning and code +2
SasAgent: Multi-Agent AI System for Small-Angle Scattering Data AnalysisLLM agents and tool use → Web, search, and research agents
Bootstrapping Reinforcement Learning with Sub-optimal Policies for Autonomous DrivingEnd-to-end autonomous driving → Reinforcement learning for driving +1
Guideline-Consistent Segmentation via Multi-Agent RefinementLLM agents and tool use → Multimodal and embodied agents
VCMamba: Bridging Convolutions with Multi-Directional Mamba for Efficient Visual RepresentationState-space sequence models (S4, Mamba) → Vision Mamba backbones
Flexible inference for animal learning rules using neural networksDeep reinforcement learning methods → Deep RL algorithms and representations
Maestro: Joint Graph & Config Optimization for Reliable AI AgentsLLM agents and tool use → General LLM agent frameworks and benchmarks
The Physical Basis of Prediction: World Model Formation in Neural Organoids via an LLM-Generated CurriculumWorld models → World models for navigation
Action Chunking with Transformers for Image-Based Spacecraft Guidance and ControlOffline RL and imitation learning from data → Imitation learning methods
Breaking to Build: A Threat Model of Prompt-Based Attacks for Securing LLMsLLM alignment / RLHF / preference optimisation → Safety alignment and jailbreak attacks +1
DisPatch: Disarming Adversarial Patches in Object Detection with Diffusion ModelsDiffusion models → Robustness, privacy and adversarial aspects
Bootstrapping Task Spaces for Self-ImprovementSynthetic data and self-improvement for LLMs → Self-evolving LLM agents
PromptEnhancer: A Simple Approach to Enhance Text-to-Image Models via Chain-of-Thought Prompt RewritingLLM reasoning → Multimodal reasoning +1
Emergent Social Dynamics of LLM Agents in the El Farol Bar ProblemLLM agents and tool use → Social simulation and multi-agent debate
Plot'n Polish: Zero-shot Story Visualization and Disentangled Editing with Text-to-Image Diffusion ModelsDiffusion models → Image generation, text-to-image and editing
EMMA: Scaling Mobile Manipulation via Egocentric Human DataOffline RL and imitation learning from data → Robot manipulation imitation learning
ArcMemo: Abstract Reasoning Composition with Lifelong LLM MemoryLLM reasoning → Multimodal reasoning
Can Language Models Handle a Non-Gregorian Calendar? The Case of the Japanese warekiLLM reasoning → Knowledge- and tool-augmented reasoning
Few-step Flow for 3D Generation via Marginal-Data Transport DistillationDiffusion models → Flow matching, time series and science
Self-adaptive Dataset Construction for Real-World Multimodal Safety ScenariosMultimodal LLMs / vision-language models → MLLM jailbreaks and adversarial robustness
Transition Models: Rethinking the Generative Learning ObjectiveDiffusion models → Diffusion foundations and sampling
SSGaussian: Semantic-Aware and Structure-Preserving 3D Style TransferDiffusion models → Image generation, text-to-image and editing
Aesthetic Image Captioning with Saliency Enhanced MLLMsMultimodal LLMs / vision-language models → Image editing and quality assessment
PagedEviction: Structured Block-wise KV Cache Pruning for Efficient Large Language Model InferenceEfficient LLM inference and serving → KV cache optimization
Connections between reinforcement learning with feedback,test-time scaling, and diffusion guidance: An anthologyLLM reasoning → Reinforcement learning for reasoning +2
Psychologically Enhanced AI AgentsLLM agents and tool use → Social simulation and multi-agent debate
From Editor to Dense Geometry EstimatorDiffusion models → 3D, point cloud and scene generation
GeoArena: Evaluating Open-World Geographic Reasoning in Large Vision-Language ModelsLLM reasoning → Spatial and embodied reasoning +1
EvoEmo: Towards Evolved Emotional Policies for Adversarial LLM Agents in Multi-Turn Price NegotiationLLM reasoning → Audio-language model reasoning +1
HumAIne-Chatbot: Real-Time Personalized Conversational AI via Reinforcement LearningLLM alignment / RLHF / preference optimisation → Human values and AI feedback alignment
Reinforcement Learning for Robust Ageing-Aware Control of Li-ion Battery Systems with Data-Driven Formal VerificationDeep reinforcement learning methods → RL for energy, networks and operations
TauGenNet: Plasma-Driven Tau PET Image Synthesis via Text-Guided 3D Diffusion ModelsDiffusion models → Medical imaging and segmentation
RL's Razor: Why Online Reinforcement Learning Forgets LessLLM reasoning → Reinforcement learning for reasoning
Learning Active Perception via Self-Evolving Preference Optimization for GUI GroundingMultimodal LLMs / vision-language models → Embodied and 3D vision-language models +2
LISM: Long-range Integrative State space Models via Input-Latent State InteractionsState-space sequence models (S4, Mamba) → Core SSM architectures for language
Are LLM Agents the New RPA? A Comparative Study with RPA Across Enterprise WorkflowsLLM agents and tool use → Software engineering coding agents
MAGneT: Coordinated Multi-Agent Generation of Synthetic Multi-Turn Mental Health Counseling SessionsLLM agents and tool use → Medical and scientific domain agents
Privacy Risks in Time Series Forecasting: User- and Record-Level Membership InferenceLLM safety and security → Membership inference attacks
Wav2DF-TSL: Two-stage Learning with Efficient Pre-training and Hierarchical Experts Fusion for Robust Audio Deepfake DetectionMixture-of-experts → Multimodal MoE models
TAGAL: Tabular Data Generation using Agentic LLM MethodsLLM agents and tool use → Web, search, and research agents
Hyper Diffusion Avatars: Dynamic Human Avatar Generation using Network Weight Space DiffusionDiffusion models → 3D, point cloud and scene generation
MEPG:Multi-Expert Planning and Generation for Compositionally-Rich Image GenerationDiffusion models → Image generation, text-to-image and editing
Towards Stable and Personalised Profiles for Lexical Alignment in Spoken Human-Agent DialogueLLM agents and tool use → General LLM agent frameworks and benchmarks
Hybrid Reinforcement Learning and Search for Flight Trajectory PlanningDeep reinforcement learning methods → RL for robot navigation and UAV control
Intermediate Languages Matter: Formal Languages and LLMs affect Neurosymbolic ReasoningLLM reasoning → Reasoning benchmarks and evaluation
Solving Robotics Tasks with Prior Demonstration via Exploration-Efficient Deep Reinforcement LearningDeep reinforcement learning methods → RL for robot locomotion and manipulation
Balancing Signal and Variance: Adaptive Offline RL Post-Training for VLA Flow ModelsVision-language-action models → Efficient VLA inference and adaptation +2
Towards an AI Musician: Synthesizing Sheet Music Problems for Musical ReasoningLLM reasoning → Audio-language model reasoning
SMooGPT: Stylized Motion Generation using Large Language ModelsDiffusion models → Video generation
Why Does Reasoning Length Converge? Unveiling the Underfitting-Overfitting Trade-off in Chain-of-ThoughtLLM reasoning → Prompting and chain-of-thought reasoning
FPC-VLA: A Vision-Language-Action Framework with a Supervisor for Failure Prediction and CorrectionVision-language-action models → Failure recovery and robustness
RTQA : Recursive Thinking for Complex Temporal Knowledge Graph Question Answering with Large Language ModelsLLM reasoning → Knowledge- and tool-augmented reasoning
Meta-Policy Reflexion: Reusable Reflective Memory and Rule Admissibility for Resource-Efficient LLM AgentLLM agents and tool use → Agent memory systems
Promptception: How Sensitive Are Large Multimodal Models to Prompts?Multimodal LLMs / vision-language models → MLLM training and visual instruction tuning
NeuroBreak: Unveil Internal Jailbreak Mechanisms in Large Language ModelsLLM alignment / RLHF / preference optimisation → Safety alignment and jailbreak attacks +1
World Model Implanting for Test-time Adaptation of Embodied AgentsWorld models → World models for language agents +1
ANTS: Adaptive Negative Textual Space Shaping for OOD Detection via Test-Time MLLM Understanding and ReasoningLLM reasoning → Multimodal reasoning
VoxRole: A Comprehensive Benchmark for Evaluating Speech-Based Role-Playing AgentsLLM agents and tool use → Social simulation and multi-agent debate
SPFT-SQL: Enhancing Large Language Model for Text-to-SQL Parsing by Self-Play Fine-TuningSynthetic data and self-improvement for LLMs → Synthetic reasoning data and bootstrapping
SelfAug: Mitigating Catastrophic Forgetting in Retrieval-Augmented Generation via Distribution Self-AlignmentRetrieval-augmented generation → RAG pipeline design and evaluation
Chain or tree? Re-evaluating complex reasoning from the perspective of a matrix of thoughtLLM reasoning → Prompting and chain-of-thought reasoning
Toward Clinically Explainable AI for Medical Diagnosis: A Foundation Model with Human-Compatible Reasoning via Reinforcement LearningLLM reasoning → Clinical and biomedical applications
Diffusion Generative Models Meet Compressed Sensing, with Applications to Imaging and FinanceDiffusion models → Diffusion foundations and sampling
Attn-Adapter: Attention Is All You Need for Online Few-shot Learner of Vision-Language ModelMultimodal LLMs / vision-language models → Prompt learning and adaptation of CLIP
Weakly-Supervised Learning of Dense Functional CorrespondencesMultimodal LLMs / vision-language models → VLM benchmarks and visual reasoning
MobileRAG: Enhancing Mobile Agent with Retrieval-Augmented GenerationRetrieval-augmented generation → Agentic RAG, agents and memory +1
FaMA: LLM-Empowered Agentic Assistant for Consumer-to-Consumer MarketplaceLLM agents and tool use → Social simulation and multi-agent debate
OccTENS: 3D Occupancy World Model via Temporal Next-Scale PredictionWorld models → World models for autonomous driving
A Comprehensive Survey on Trustworthiness in Reasoning with Large Language ModelsLLM reasoning → Reasoning for agents, planning and code
Data-Augmented Quantization-Aware Knowledge DistillationKnowledge distillation → Knowledge distillation for LLMs
Meta-Inverse Reinforcement Learning for Mean Field Games via Probabilistic Context VariablesOffline RL and imitation learning from data → Inverse reinforcement learning
Vehicle-to-Infrastructure Collaborative Spatial Perception via Multimodal Large Language ModelsMultimodal LLMs / vision-language models → Driving VLMs and open-vocabulary detection
An Agentic Model Context Protocol Framework for Medical Concept StandardizationLLM agents and tool use → Medical and scientific domain agents
Learning to Deliberate: Meta-policy Collaboration for Agentic LLMs with Multi-agent Reinforcement LearningLLM reasoning → Reasoning for agents, planning and code +1
Align-then-Slide: A complete evaluation framework for Ultra-Long Document-Level Machine TranslationLLM alignment / RLHF / preference optimisation → LLM preference optimization methods
Measuring How (Not Just Whether) VLMs Build Common GroundLLM reasoning → Multimodal reasoning +1
Causality-guided Prompt Learning for Vision-language Models via Visual GranulationMultimodal LLMs / vision-language models → Prompt learning and adaptation of CLIP
MedVista3D: Vision-Language Modeling for Reducing Diagnostic Errors in 3D CT Disease Detection, Understanding and ReportingMultimodal LLMs / vision-language models → Medical vision-language models
Fitting Image Diffusion Models on Video DatasetsDiffusion models → Video generation
SAMVAD: A Multi-Agent System for Simulating Judicial Deliberation Dynamics in IndiaLLM agents and tool use → Social simulation and multi-agent debate
What Fundamental Structure in Reward Functions Enables Efficient Sparse-Reward Learning?Offline RL and imitation learning from data → Inverse reinforcement learning
Evaluating the Robustness of Retrieval-Augmented Generation to Adversarial Evidence in the Health DomainRetrieval-augmented generation → Medical and biomedical RAG
Iti-Validator: A Guardrail Framework for Validating and Correcting LLM-Generated Itineraries
STM-Graph: A Python Framework for Spatio-Temporal Mapping and Graph Neural Network Predictions
An Interpretable Ensemble Framework for Multi-Omics Dementia Biomarker Discovery Under HDLSS Conditions
Data-Efficient Psychiatric Disorder Detection via Self-supervised Learning on Frequency-enhanced Brain Networks
From Predictions to Explanations: Explainable AI for Autism Diagnosis and Identification of Critical Brain Regions
Multimodal Deep Learning for ATCO Command Lifecycle Modeling and Workload Prediction
Variational Gaussian Mixture Manifold Models for Client-Specific Federated Personalization

All 304