Papers

New on arXiv

Tue 23 Dec 2025

282 papers

Domain-Partitioned Hybrid RAG for Legal Reasoning: Toward Modular and Explainable Legal AI for IndiaLLM reasoning → Reasoning for agents, planning and code +1
Energy-Efficient Omnidirectional Locomotion for Wheeled Quadrupeds via Predictive Energy-Aware Nominal Gait SelectionDeep reinforcement learning methods → RL for robot locomotion and manipulation
Energy-Aware Routing to Large Reasoning ModelsLLM reasoning → Test-time compute and efficient thinking
Calibrating LLM Judges: Linear Probes for Fast and Reliable Uncertainty EstimationMechanistic interpretability → Probing deception, bias and beliefs
Multi-objective hybrid knowledge distillation for efficient deep learning in smart agricultureKnowledge distillation → Detection and dense prediction distillation
Masking Teacher and Reinforcing Student for Distilling Vision-Language ModelsMultimodal LLMs / vision-language models → MLLM training and visual instruction tuning +1
Meta-information Guided Cross-domain Synergistic Diffusion Model for Low-dose PET ReconstructionDiffusion models → Medical imaging and segmentation
DiRL: An Efficient Post-Training Framework for Diffusion Language ModelsDiffusion models → Discrete and text diffusion
VideoScaffold: Elastic-Scale Visual Hierarchies for Streaming Video Understanding in MLLMsMultimodal LLMs / vision-language models → Video understanding with MLLMs
From Visual Perception to Deep Empathy: An Automated Assessment Framework for House-Tree-Person Drawings Using Multimodal LLMs and Multi-Agent CollaborationMultimodal LLMs / vision-language models → Hallucination mitigation in LVLMs +1
Nemotron 3 Nano: Open, Efficient Mixture-of-Experts Hybrid Mamba-Transformer Model for Agentic ReasoningLLM reasoning → Test-time compute and efficient thinking +3
MAR:Multi-Agent Reflexion Improves Reasoning Abilities in LLMsLLM reasoning → Reasoning for agents, planning and code +1
Input-Adaptive Visual Preprocessing for Efficient Fast Vision-Language Model InferenceLLM reasoning → Multimodal reasoning +1
Context-Sensitive Abstractions for Reinforcement Learning with Parameterized ActionsDeep reinforcement learning methods → Deep RL algorithms and representations
MediEval: A Unified Medical Benchmark for Patient-Contextual and Knowledge-Grounded Reasoning in LLMsLLM reasoning → Clinical and biomedical applications
Divided We Fall: Defending Against Adversarial Attacks via Soft-Gated Fractional Mixture-of-Experts with Randomized Adversarial TrainingMixture-of-experts → Vision mixture-of-experts
Learning to Sense for Driving: Joint Optics-Sensor-Model Co-Design for Semantic SegmentationEnd-to-end autonomous driving → VLMs for autonomous driving
Semantic Deception: When Reasoning Models Can't Compute an AdditionLLM reasoning → Reasoning benchmarks and evaluation
Safety Alignment of LMs via Non-cooperative GamesLLM alignment / RLHF / preference optimisation → Safety alignment and jailbreak attacks +1
A Benchmark for Evaluating Outcome-Driven Constraint Violations in Autonomous AI AgentsLLM agents and tool use → Agent threats, guardrails and governance
Investigating Model Editing for Unlearning in Large Language ModelsLLM safety and security → Machine unlearning
X-GridAgent: An LLM-Powered Agentic AI System for Assisting Power Grid AnalysisLLM agents and tool use → Financial trading and forecasting agents
Soft Filtering: Guiding Zero-shot Composed Image Retrieval with Prescriptive and Proscriptive ConstraintsMultimodal LLMs / vision-language models → Document understanding and multimodal RAG
Generalization of RLVR Using Causal Reasoning as a TestbedLLM reasoning → Reinforcement learning for reasoning
Bridging Efficiency and Safety: Formal Verification of Neural Networks with Early ExitsEfficient LLM inference and serving → Early exit and edge inference
AgentMath: Empowering Mathematical Reasoning for Large Language Models via Tool-Augmented AgentLLM reasoning → Reasoning for agents, planning and code +1
VL4Gaze: Unleashing Vision-Language Models for Gaze FollowingMultimodal LLMs / vision-language models → VLM benchmarks and visual reasoning
FEM-Bench: A Structured Scientific Reasoning Benchmark for Evaluating Code-Generating LLMsLLM reasoning → Spatial and embodied reasoning
SemanticGen: Video Generation in Semantic SpaceDiffusion models → Video generation
LongVideoAgent: Multi-Agent Reasoning with Long VideosLLM reasoning → Multimodal reasoning +2
Active Intelligence in Video Avatars via Closed-loop World ModelingWorld models → Video-generation world models and simulators
Probing and Leveraging Video Diffusion Transformer Features for Robust Point TrackingDiffusion models → Video generation
Emergent temporal abstractions in autoregressive models enable hierarchical reinforcement learningDeep reinforcement learning methods → Deep RL algorithms and representations
MoE-DiffuSeq: Enhancing Long-Document Diffusion Models with Sparse Attention and Mixture of ExpertsDiffusion models → Discrete and text diffusion +1
Cube Bench: A Benchmark for Spatial Visual Reasoning in MLLMsLLM reasoning → Spatial and embodied reasoning +1
Leveraging High-Fidelity Digital Models and Reinforcement Learning for Mission Engineering: A Case Study of Aerial Firefighting Under Perfect InformationDeep reinforcement learning methods → RL for robot navigation and UAV control
Automated stereotactic radiosurgery planning using a human-in-the-loop reasoning large language model agentLLM reasoning → Clinical and biomedical applications +1
Performative Policy Gradient: Optimality in Performative Reinforcement LearningDeep reinforcement learning methods → RL theory and sample complexity
Fail Fast, Win Big: Rethinking the Drafting Strategy in Speculative Decoding via Diffusion LLMsLLM reasoning → Test-time compute and efficient thinking +1
LEAD: Minimizing Learner-Expert Asymmetry in End-to-End DrivingEnd-to-end autonomous driving → Imitation-learning end-to-end planners +1
FlashVLM: Text-Guided Visual Token Selection for Large Multimodal ModelsMultimodal LLMs / vision-language models → Visual token pruning and efficiency
Learning to Reason in 4D: Dynamic Spatial Understanding for Vision Language ModelsLLM reasoning → Spatial and embodied reasoning +1
ARBITER: AI-Driven Filtering for Role-Based Access ControlRetrieval-augmented generation → Privacy and security of RAG
Recurrent Off-Policy Deep Reinforcement Learning Doesn't Have to be SlowDeep reinforcement learning methods → Deep RL algorithms and representations
Bohrium + SciMaster: Building the Infrastructure and Ecosystem for Agentic Science at ScaleLLM agents and tool use → Web, search, and research agents
Laser: Governing Long-Horizon Agentic Search via Structured Protocol and Context RegisterLLM reasoning → Reasoning for agents, planning and code +1
Chain-of-Anomaly Thoughts with Large Vision-Language ModelsLLM reasoning → Chain-of-thought reasoning for driving +1
ChatGPT: Excellent Paper! Accept It. Editor: Imposter Found! Review RejectedLLM safety and security → Prompt injection and agent security
Identifying Appropriately-Sized Services with Deep Reinforcement LearningDeep reinforcement learning methods → RL for energy, networks and operations
CRAFT: Continuous Reasoning and Agentic Feedback Tuning for Multimodal Text-to-Image GenerationLLM reasoning → Multimodal reasoning +1
The devil is in the details: Enhancing Video Virtual Try-On via Keyframe-Driven Details InjectionDiffusion models → 3D, point cloud and scene generation
Branch Learning in MRI: More Data, More Models, More TrainingMixture-of-experts → Medical and biomedical MoE
Can LLMs Solve My Grandma's Riddle? Evaluating Multilingual Large Language Models on Reasoning Traditional Bangla Tricky RiddlesLLM reasoning → Reasoning benchmarks and evaluation
LP-CFM: Perceptual Invariance-Aware Conditional Flow Matching for Speech ModelingDiffusion models → Audio, speech and music
TableGPT-R1: Advancing Tabular Reasoning Through Reinforcement LearningLLM reasoning → Knowledge- and tool-augmented reasoning
Structured Visualization Design Knowledge for Grounding Generative Reasoning and Situated FeedbackLLM reasoning → Multimodal reasoning
AprielGuardLLM agents and tool use → Agent threats, guardrails and governance +1
Mixture-of-Experts with Gradient Conflict-Driven Subspace Topology Pruning for Emergent ModularityMixture-of-experts → Expert routing strategies
Synthesizing Procedural Memory: Challenges and Architectures in Automated Workflow GenerationLLM agents and tool use → Agent memory systems
ActionFlow: A Pipelined Action Acceleration for Vision Language Models on EdgeMultimodal LLMs / vision-language models → Embodied and 3D vision-language models +1
Graph-Symbolic Policy Enforcement and Control (G-SPEC): A Neuro-Symbolic Framework for Safe Agentic AI in 5G Autonomous NetworksLLM agents and tool use → Agentic AI for wireless networks
MemR3: Memory Retrieval via Reflective Reasoning for LLM AgentsLLM reasoning → Reasoning for agents, planning and code +1
How I Met Your Bias: Investigating Bias Amplification in Diffusion ModelsDiffusion models → Diffusion foundations and sampling
Generalisation in Multitask Fitted Q-Iteration and Offline Q-learningDeep reinforcement learning methods → Deep RL algorithms and representations
Predictive-LoRA: A Proactive and Fragmentation-Aware Serverless Inference System for LLMsEfficient LLM inference and serving → Serving systems and request scheduling
TongSIM: A General Platform for Simulating Intelligent MachinesMultimodal LLMs / vision-language models → Embodied and 3D vision-language models
Asynchronous Fast-Slow Vision-Language-Action Policies for Whole-Body Robotic ManipulationLLM reasoning → Spatial and embodied reasoning +2
FaithLens: Detecting and Explaining Faithfulness HallucinationRetrieval-augmented generation → Hallucination detection in RAG
Offline Safe Policy Optimization From Heterogeneous FeedbackLLM alignment / RLHF / preference optimisation → Reward modeling and RLHF
Learning to Reason in LLMs by Expectation MaximizationLLM reasoning → Test-time compute and efficient thinking
Odysseus: Jailbreaking Commercial Multimodal LLM-integrated Systems via Dual SteganographyMultimodal LLMs / vision-language models → MLLM jailbreaks and adversarial robustness +1
SigLino: Efficient Multi-Teacher Distillation for Agglomerative Vision Foundation ModelsMixture-of-experts → Vision mixture-of-experts
CoDi -- an exemplar-conditioned diffusion model for low-shot countingDiffusion models → Image generation, text-to-image and editing
Retrieval-augmented Prompt Learning for Pre-trained Foundation ModelsRetrieval-augmented generation → Generation and context integration
Multi-hop Reasoning via Early Knowledge AlignmentLLM reasoning → Knowledge- and tool-augmented reasoning +1
M3KG-RAG: Multi-hop Multimodal Knowledge Graph-enhanced Retrieval-Augmented GenerationMultimodal LLMs / vision-language models → Document understanding and multimodal RAG +1
MolAct: An Agentic RL Framework for Molecular Editing and Property OptimizationLLM agents and tool use → Medical and scientific domain agents
milliMamba: Specular-Aware Human Pose Estimation via Dual mmWave Radar with Multi-Frame Mamba FusionState-space sequence models (S4, Mamba) → Video, motion and generative Mamba models
HEART-VIT: Hessian-Guided Efficient Dynamic Attention and Token Pruning in Vision TransformerEfficient LLM inference and serving → Token pruning and merging for ViTs
Sample-Efficient Policy Constraint Offline Deep Reinforcement Learning based on Sample FilteringOffline RL and imitation learning from data → Offline RL algorithms
Generative Spectrum Cartography: Unified Reconstruction and Active Sensing via Diffusion ModelsDiffusion models → Inverse problems and image restoration
LiDARDraft: Generating LiDAR Point Cloud from Versatile InputsWorld models → World modeling from LiDAR generation
Memory-T1: Reinforcement Learning for Temporal Reasoning in Multi-session AgentsLLM reasoning → Reasoning for agents, planning and code
Reason2Decide: Rationale-Driven Multi-Task LearningLLM reasoning → Clinical and biomedical applications
An Optimal Policy for Learning Controllable Dynamics by ExplorationDeep reinforcement learning methods → RL theory and sample complexity
Bring My Cup! Personalizing Vision-Language-Action Models with Visual Attentive PromptingVision-language-action models → Reasoning and chain-of-thought in VLA
Control Variate Score Matching for Diffusion ModelsDiffusion models → Diffusion foundations and sampling
Few-Shot-Based Modular Image-to-Video Adapter for Diffusion ModelsDiffusion models → Video generation
Schoenfeld's Anatomy of Mathematical Reasoning by Language ModelsLLM reasoning → Mathematical reasoning
Toward General Semantic Chunking: A Discriminative Framework for Ultra-Long Documents
From Textbook to Talkbot: A Case Study of a Greek-Language RAG-Based Chatbot in Higher Education
DeepEvidence: Empowering Biomedical Discovery with Deep Knowledge Graph Research
Multi-Level Embedding Conformer Framework for Bengali Automatic Speech Recognition
M'as contexto no es mejor. Paradoja de la diluci'on vectorial en RAG corporativos
AI Application Operations -- A Socio-Technical Framework for Data-driven Organizations
Why Slop Matters
Context Video Semantic Transmission with Variable Length and Rate Coding over MIMO Channels
Base Station Deployment under EMF constrain by Deep Reinforcement learning
Fidelity Before Structure: Verbatim Chunks Beat Lossy Artifact Extraction in Long-Conversation LLM Memory
Predicting Mycotoxin Contamination in Irish Oats Using Deep and Transfer Learning

All 282