Papers

New on arXiv

Thu 16 Oct 2025

426 papers

Generative human motion mimicking through feature extraction in denoising diffusion settingsDiffusion models → Video generation
Sketch2BIM: A Multi-Agent Human-AI Collaborative Pipeline to Convert Hand-Drawn Floor Plans to 3D BIMLLM agents and tool use → Multimodal and embodied agents
Benchmarking Reasoning Reliability in Artificial Intelligence Models for Energy-System AnalysisLLM reasoning → Test-time compute and efficient thinking
UniCode: Augmenting Evaluation for Code ReasoningLLM reasoning → Reasoning for agents, planning and code
Privacy-R1: Privacy-Aware Multi-LLM Agent Collaboration via Reinforcement LearningLLM agents and tool use → Agent threats, guardrails and governance
Kelle: Co-design KV Caching and eDRAM for Efficient LLM Serving in Edge ComputingEfficient LLM inference and serving → KV cache optimization
Vector Quantization in the Brain: Grid-like Codes in World ModelsWorld models → Latent predictive world models
Membership Inference over Diffusion-models-based Synthetic Tabular DataDiffusion models → Robustness, privacy and adversarial aspects +1
IAD-GPT: Advancing Visual Knowledge in Multimodal Large Language Model for Industrial Anomaly DetectionMultimodal LLMs / vision-language models → Chart understanding and anomaly detection
RoBCtrl: Attacking GNN-Based Social Bot Detectors via Reinforced Manipulation of Bots Control InteractionDeep reinforcement learning methods → Adversarial robustness and RL in finance
Disaster Management in the Era of Agentic AI Systems: A Vision for Collective Human-Machine Intelligence for Augmented ResilienceLLM agents and tool use → Multimodal and embodied agents
Structure-R1: Dynamically Leveraging Structural Knowledge in LLM Reasoning through Reinforcement LearningLLM reasoning → Knowledge- and tool-augmented reasoning +1
RM-RL: Role-Model Reinforcement Learning for Precise Robot ManipulationOffline RL and imitation learning from data → Real-world robot offline RL +1
MAGPIE: A benchmark for Multi-AGent contextual PrIvacy EvaluationLLM agents and tool use → Agent threats, guardrails and governance
Policy Transfer for Continuous-Time Reinforcement Learning: A (Rough) Differential Equation ApproachDeep reinforcement learning methods → RL theory and sample complexity
Train a Unified Multimodal Data Quality Classifier with Synthetic DataMultimodal LLMs / vision-language models → VLM benchmarks and visual reasoning +1
HugAgent: A Human Simulation Benchmark for Individual-Level ReasoningLLM reasoning → Reasoning benchmarks and evaluation
Inferring Relative Consequences of Mechanical Ventilation from Observational Data Using Game-Based ComparisonsDeep reinforcement learning methods → RL for clinical treatment and surgery
Procedural Game Level Design with Deep Reinforcement LearningDeep reinforcement learning methods → RL for games and combinatorial optimization
Deep generative priors for 3D brain analysisDiffusion models → Medical imaging and segmentation
DLER: Doing Length pEnalty Right - Incentivizing More Intelligence per Token via Reinforcement LearningLLM reasoning → Test-time compute and efficient thinking +1
Targeted Attacks and Defenses for Distributed Federated Learning in Vehicular NetworksLLM safety and security → Federated unlearning and poisoning
PoTS: Proof-of-Training-Steps for Backdoor Detection in Large Language ModelsLLM safety and security → Backdoor and data poisoning attacks
Operator Flow Matching for Timeseries ForecastingDiffusion models → Flow matching, time series and science
OpenEstimate: Evaluating LLMs on Reasoning Under Uncertainty with Real-World DataLLM reasoning → Test-time compute and efficient thinking
Assessing Coherency and Consistency of Code Execution Reasoning by Large Language ModelsLLM reasoning → Reasoning for agents, planning and code
Sequential Comics for Jailbreaking Multimodal Large Language Models via Structured Visual StorytellingMultimodal LLMs / vision-language models → MLLM jailbreaks and adversarial robustness +1
Learn to Change the World: Multi-level Reinforcement Learning with Model-Changing ActionsDeep reinforcement learning methods → RL theory and sample complexity
DRIFT: Transferring Reasoning Priors for Efficient MLLM Fine-TuningLLM reasoning → Multimodal reasoning +1
Why Do LLM Agents Fail in Exploring New Environments? A World-Modeling PerspectiveWorld models → World models for language agents +1
Comprehensive language-image pre-training for 3D medical image understandingMultimodal LLMs / vision-language models → Medical vision-language models
Generalized Dynamics Generation towards Scannable Physical World ModelWorld models → Latent predictive world models
Composition-Grounded Data Synthesis for Visual ReasoningLLM reasoning → Multimodal reasoning +2
LoRAverse: A Submodular Framework to Retrieve Diverse Adapters for Diffusion ModelsDiffusion models → Image generation, text-to-image and editing
Active Honeypot Guardrail System: Probing and Confirming Multi-Turn LLM JailbreaksLLM safety and security → Jailbreaking
Rethinking Toxicity Evaluation in Large Language Models: A Multi-Label PerspectiveLLM reasoning → Reasoning benchmarks and evaluation +1
ES-C51: Expected Sarsa Based C51 Distributional Reinforcement Learning AlgorithmDeep reinforcement learning methods → RL theory and sample complexity
Coupled Diffusion Sampling for Training-Free Multi-View Image EditingDiffusion models → 3D, point cloud and scene generation
Compositional Machine Design as Program Synthesis with LLMsLLM agents and tool use → General LLM agent frameworks and benchmarks
From Pixels to Words -- Towards Native Vision-Language Primitives at ScaleMultimodal LLMs / vision-language models → VLM benchmarks and visual reasoning
Learning an Image Editing Model without Image Editing PairsDiffusion models → Image generation, text-to-image and editing
Terra: Explorable Native 3D World Model with Point LatentsWorld models → Video-generation world models and simulators
Attention Is All You Need for KV Cache in Diffusion LLMsEfficient LLM inference and serving → Diffusion LLM and MoE inference
RDD: Retrieval-Based Demonstration Decomposer for Planner Alignment in Long-Horizon TasksMultimodal LLMs / vision-language models → Embodied and 3D vision-language models +1
Information Gain-based Policy Optimization: A Simple and Effective Approach for Multi-Turn Search AgentsLLM agents and tool use → Reinforcement learning for LLM agents
Efficient Parallel Samplers for Recurrent-Depth Models and Their Connection to Diffusion Language ModelsDiffusion models → Discrete and text diffusion
CBF-RL: Safety Filtering Reinforcement Learning in Training with Control Barrier FunctionsDeep reinforcement learning methods → Safe and constrained RL
MathCanvas: Intrinsic Visual Chain-of-Thought for Multimodal Mathematical ReasoningLLM reasoning → Multimodal reasoning
RealDPO: Real or Not Real, that is the PreferenceLLM alignment / RLHF / preference optimisation → Diffusion and multimodal model alignment
EdgeNavMamba: Mamba Optimized Object Detection for Energy Efficient Edge DevicesState-space sequence models (S4, Mamba) → Point clouds, detection and remote sensing
LaSeR: Reinforcement Learning with Last-Token Self-RewardingLLM reasoning → Reinforcement learning for reasoning +2
GroundedPRM: Tree-Guided and Fidelity-Aware Process Reward Modeling for Step-Level ReasoningLLM reasoning → Reinforcement learning for reasoning +1
Circuit Insights: Towards Interpretability Beyond ActivationsMechanistic interpretability → Circuits and mechanistic analysis
VT-Refine: Learning Bimanual Assembly with Visuo-Tactile Feedback via Simulation Fine-TuningOffline RL and imitation learning from data → Robot manipulation imitation learning +1
Harmonizing Diverse Models: A Layer-wise Merging Strategy for Consistent GenerationRetrieval-augmented generation → RAG pipeline design and evaluation
Budget-aware Test-time Scaling via Discriminative VerificationLLM reasoning → Test-time compute and efficient thinking
VLA^2: Empowering Vision-Language-Action Models with an Agentic Framework for Unseen Concept ManipulationVision-language-action models → Generalist VLA models and benchmarks +1
Reasoning with Sampling: Your Base Model is Smarter Than You ThinkLLM reasoning → Reinforcement learning for reasoning
Mapping Smarter, Not Harder: A Test-Time Reinforcement Learning Agent That Improves Without Labels or Model UpdatesLLM agents and tool use → Web, search, and research agents +1
Convergence of an actor-critic gradient flow for entropy regularised MDPs in general spacesDeep reinforcement learning methods → RL theory and sample complexity
Leveraging Multimodal LLM Descriptions of Activity for Explainable Semi-Supervised Video Anomaly DetectionMultimodal LLMs / vision-language models → Chart understanding and anomaly detection
You May Speak Freely: Improving the Fine-Grained Visual Recognition Capabilities of Multimodal Large Language Models with Answer ExtractionMultimodal LLMs / vision-language models → VLM benchmarks and visual reasoning
Benchmarking Multimodal Large Language Models for Face RecognitionMultimodal LLMs / vision-language models → Social bias and fairness in VLMs
Rewiring Experts on the Fly:Continuous Rerouting for Better Online Adaptation in Mixture-of-Expert modelsMixture-of-experts → Expert routing strategies
Where to Search: Measure the Prior-Structured Search Space of LLM AgentsLLM reasoning → Reasoning for agents, planning and code +1
Provable Unlearning with Gradient Ascent on Two-Layer ReLU Neural NetworksLLM safety and security → Machine unlearning
Reinforcement Learning with Stochastic Reward MachinesDeep reinforcement learning methods → Interpretable and logic-guided RL
QDepth-VLA: Quantized Depth Prediction as Auxiliary Supervision for Vision-Language-Action ModelsVision-language-action models → 3D and spatial-aware VLA
RL-100: Performant Robotic Manipulation with Real-World Reinforcement LearningDeep reinforcement learning methods → RL for robot locomotion and manipulation
RoboGPT-R1: Enhancing Robot Task Planning with Reinforcement LearningLLM reasoning → Spatial and embodied reasoning +2
Supervised Fine-Tuning or Contrastive Learning? Towards Better Multimodal LLM RerankingMultimodal LLMs / vision-language models → Document understanding and multimodal RAG
Agentic NL2SQL to Reduce Computational CostsLLM agents and tool use → Web, search, and research agents
SkyDreamer: Interpretable End-to-End Vision-Based Drone Racing with Model-Based Reinforcement LearningWorld models → World models for navigation +1
Finding Answers in Thought Matters: Revisiting Evaluation on Large Language Models with ReasoningLLM reasoning → Reasoning benchmarks and evaluation
Open TeleDex: A Hardware-Agnostic Teleoperation System for Imitation Learning based Dexterous ManipulationOffline RL and imitation learning from data → Robot manipulation imitation learning
Inpainting the Red Planet: Diffusion Models for the Reconstruction of Martian Environments in Virtual RealityDiffusion models → 3D, point cloud and scene generation
COIG-Writer: A High-Quality Dataset for Chinese Creative Writing with Thought ProcessesLLM reasoning → Reasoning benchmarks and evaluation
DEXTER: Diffusion-Guided EXplanations with TExtual Reasoning for Vision ModelsLLM reasoning → Multimodal reasoning
AutoRubric: Rubric-Based Generative Rewards for Faithful Multimodal ReasoningLLM reasoning → Multimodal reasoning +1
The Pursuit of Diversity: Multi-Objective Testing of Deep Reinforcement Learning AgentsDeep reinforcement learning methods → RL for autonomous driving and traffic
ToolPRM: Fine-Grained Inference Scaling of Structured Outputs for Function CallingLLM agents and tool use → Tool learning and function calling
Cognitive-Aligned Spatio-Temporal Large Language Models For Next Point-of-Interest PredictionLLM alignment / RLHF / preference optimisation → Preference alignment for recommenders and RAG
LLM Agents for Automated Web Vulnerability Reproduction: Are We There Yet?LLM agents and tool use → Agent threats, guardrails and governance
xLLM Technical ReportEfficient LLM inference and serving → Serving systems and request scheduling
VTimeCoT: Thinking by Drawing for Video Temporal Grounding and ReasoningLLM reasoning → Multimodal reasoning +1
WeCKD: Weakly-supervised Chained Distillation Network for Efficient Multimodal Medical ImagingKnowledge distillation → Distillation for medical imaging
Beyond "Hallucinations": A Framework for Stable Human-AI ReasoningLLM reasoning → Reasoning benchmarks and evaluation
Generative Models From and For Sampling-Based MPC: A Bootstrapped Approach For Adaptive Contact-Rich ManipulationDiffusion models → Diffusion and flow policies for robotics
The Bidding Games: Reinforcement Learning for MEV Extraction on Polygon BlockchainDeep reinforcement learning methods → Adversarial robustness and RL in finance
MR.Rec: Synergizing Memory and Reasoning for Personalized Recommendation Assistant with LLMsLLM reasoning → LLM reasoning for recommendation +1
RLAIF-SPA: Structured AI Feedback for Semantic-Prosodic Alignment in Speech SynthesisLLM alignment / RLHF / preference optimisation → Preference alignment for recommenders and RAG
Efficient Video Sampling: Pruning Temporally Redundant Tokens for Faster VLM InferenceMultimodal LLMs / vision-language models → Visual token pruning and efficiency
LeapFactual: Reliable Visual Counterfactual Explanation Using Conditional Flow MatchingDiffusion models → Diffusion foundations and sampling
ColorBench: Benchmarking Mobile Agents with Graph-Structured Framework for Complex Long-Horizon TasksLLM agents and tool use → GUI and mobile agents
Code-driven Number Sequence Calculation: Enhancing the inductive Reasoning Abilities of Large Language ModelsLLM reasoning → Reasoning benchmarks and evaluation
Beyond Correctness: Evaluating Subjective Writing Preferences Across CulturesLLM alignment / RLHF / preference optimisation → Human values and AI feedback alignment
Accelerated Multi-Modal Motion Planning Using Context-Conditioned Diffusion ModelsDiffusion models → Diffusion and flow policies for robotics
Knowledge-based Visual Question Answer with Multimodal Processing, Retrieval and FilteringMultimodal LLMs / vision-language models → Document understanding and multimodal RAG
Multimodal RAG for Unstructured Data:Leveraging Modality-Aware Knowledge Graphs with Hybrid RetrievalRetrieval-augmented generation → Multimodal RAG
Talking Points: Describing and Localizing PixelsMultimodal LLMs / vision-language models → VLM benchmarks and visual reasoning

All 426