Papers

New on arXiv

Mon 22 Dec 2025

324 papers

PediaMind-R1: A Temperament-Aware Language Model for Personalized Early Childhood Care Reasoning via Cognitive Modeling and Preference AlignmentLLM reasoning → Clinical and biomedical applications +1
Directional Attractors in LLM Reasoning: How Similarity Retrieval Steers Iterative Summarization Based ReasoningLLM reasoning → Test-time compute and efficient thinking
A Multi-Perspective Benchmark and Moderation Model for Evaluating Safety and Adversarial RobustnessLLM safety and security → Safety alignment, refusal and guardrails
Agentic AI for Autonomous, Explainable, and Real-Time Credit Risk Decision-MakingLLM agents and tool use → Financial trading and forecasting agents
VLM-PAR: A Vision Language Model for Pedestrian Attribute RecognitionMultimodal LLMs / vision-language models → Driving VLMs and open-vocabulary detection
With Great Capabilities Come Great Responsibilities: Introducing the Agentic Risk & Capability Framework for Governing Agentic AI SystemsLLM agents and tool use → Agent threats, guardrails and governance
Super-Resolution Enhancement of Medical Images Based on Diffusion Model: An Optimization Scheme for Low-Resolution Gastric ImagesDiffusion models → Medical imaging and segmentation
GamiBench: Evaluating Spatial Reasoning and 2D-to-3D Planning Capabilities of MLLMs with Origami Folding TasksLLM reasoning → Spatial and embodied reasoning +1
Reflection-Driven Control for Trustworthy Code AgentsLLM agents and tool use → Agent threats, guardrails and governance +1
PHOTON: Hierarchical Autoregressive Modeling for Lightspeed and Memory-Efficient Language GenerationEfficient LLM inference and serving → Speculative decoding
Diffusion Models in Simulation-Based Inference: A Tutorial ReviewDiffusion models → Diffusion foundations and sampling
PRISM: A Personality-Driven Multi-Agent Framework for Social Media SimulationLLM agents and tool use → Social simulation and multi-agent debate
Widget2Code: From Visual Widgets to UI Code via Multimodal LLMsMultimodal LLMs / vision-language models → Embodied and 3D vision-language models
Demystifying LLM-as-a-Judge: Analytically Tractable Model for Inference-Time ScalingLLM reasoning → Test-time compute and efficient thinking
HARMON-E: Hierarchical Agentic Reasoning for Multimodal Oncology Notes to Extract Structured DataLLM reasoning → Clinical and biomedical applications
UCCL-EP: Portable Expert-Parallel CommunicationMixture-of-experts → Efficient MoE training and inference systems
Learning to Refocus with Video Diffusion ModelsDiffusion models → Video generation
PhysMaster: Building an Autonomous AI Physicist for Theoretical and Computational Physics ResearchLLM reasoning → Spatial and embodied reasoning +1
A Declarative Language for Building And Orchestrating LLM-Powered Agent WorkflowsLLM agents and tool use → Software engineering coding agents
Interact2Ar: Full-Body Human-Human Interaction Generation via Autoregressive Diffusion ModelsDiffusion models → Video generation
Visual-Aware CoT: Achieving High-Fidelity Visual Consistency in Unified ModelsLLM reasoning → Multimodal reasoning
From Indoor to Open World: Revealing the Spatial Reasoning Gap in MLLMsLLM reasoning → Spatial and embodied reasoning +1
GenEnv: Difficulty-Aligned Co-Evolution Between LLM Agents and Environment SimulatorsLLM agents and tool use → General LLM agent frameworks and benchmarks
VA-π: Variational Policy Alignment for Pixel-Aware Autoregressive GenerationLLM alignment / RLHF / preference optimisation → Diffusion and multimodal model alignment
WorldWarp: Propagating 3D Geometry with Asynchronous Video DiffusionDiffusion models → 3D, point cloud and scene generation
Efficient Vision Mamba for MRI Super-Resolution via Hybrid Selective ScanningState-space sequence models (S4, Mamba) → Mamba for medical image analysis
Multimodal LLMs for Historical Dataset Construction from Archival Image Scans: German Patents (1877-1918)Multimodal LLMs / vision-language models → MLLM training and visual instruction tuning
Exploring Zero-Shot ACSA with Unified Meaning Representation in Chain-of-Thought PromptingLLM reasoning → Prompting and chain-of-thought reasoning
Generative diffusion models for agricultural AI: plant image generation, indoor-to-outdoor translation, and expert preference alignmentDiffusion models → Inverse problems and image restoration
MapTrace: Scalable Data Generation for Route Tracing on MapsMultimodal LLMs / vision-language models → Embodied and 3D vision-language models
Increasing the Thinking Budget is Not All You NeedLLM reasoning → Test-time compute and efficient thinking
Patlak Parametric Image Estimation from Dynamic PET Using Diffusion Model PriorDiffusion models → Medical imaging and segmentation
The Epistemological Consequences of Large Language Models: Rethinking collective intelligence and institutional knowledgeLLM reasoning → Reasoning for agents, planning and code
Results of the 2024 CommonRoad Motion Planning Competition for Autonomous VehiclesEnd-to-end autonomous driving → End-to-end system architecture
REALM: A Real-to-Sim Validated Benchmark for Generalization in Robotic ManipulationVision-language-action models → Generalist VLA models and benchmarks
CARE What Fails: Contrastive Anchored-REflection for Verifiable Multimodal ReasoningLLM reasoning → Multimodal reasoning
StoryMem: Multi-shot Long Video Storytelling with MemoryDiffusion models → Video generation
Event Extraction in Large Language ModelMultimodal LLMs / vision-language models → Video understanding with MLLMs
CASA: Cross-Attention over Self-Attention for Efficient Vision-Language FusionMultimodal LLMs / vision-language models → Visual token pruning and efficiency
QuantiPhy: A Quantitative Benchmark Evaluating Physical Reasoning Abilities of Vision-Language ModelsLLM reasoning → Spatial and embodied reasoning +1
LacaDM: A Latent Causal Diffusion Model for Multiobjective Reinforcement LearningDiffusion models → Diffusion and flow policies for robotics +1
Anatomy-R1: Enhancing Anatomy Reasoning in Multimodal Large Language Models via Anatomical Similarity Curriculum and Group Diversity AugmentationLLM reasoning → Clinical and biomedical applications +1
Emotion-Director: Bridging Affective Shortcut in Emotion-Oriented Image GenerationDiffusion models → Audio, speech and music
VASP Agent: An Agentic Framework for Autonomous First-principles CalculationsLLM agents and tool use → Medical and scientific domain agents
MaP-AVR: A Meta-Action Planner for Agents Leveraging Vision Language Models and Retrieval-Augmented GenerationMultimodal LLMs / vision-language models → Embodied and 3D vision-language models +2
D2Pruner: Debiased Importance and Structural Diversity for MLLM Token PruningMultimodal LLMs / vision-language models → Visual token pruning and efficiency +1
Real-Time Streamable Generative Speech Restoration with Flow MatchingDiffusion models → Audio, speech and music
dMLLM-TTS: Self-Verified and Efficient Test-Time Scaling for Diffusion Multi-Modal Large Language ModelsLLM reasoning → Diffusion reasoning and test-time scaling +1
MobileWorld: Benchmarking Autonomous Mobile Agents in Agent-User Interactive and MCP-Augmented EnvironmentsLLM agents and tool use → GUI and mobile agents
From Retrieval to Reasoning: A Framework for Cyber Threat Intelligence NER with Explicit and Adaptive InstructionsLLM reasoning → Reasoning for agents, planning and code
Brain-Grounded Axes for Reading and Steering LLM StatesMechanistic interpretability → Activation steering methods
EchoTrail-GUI: Building Actionable Memory for GUI Agents via Critic-Guided Self-ExplorationMultimodal LLMs / vision-language models → Embodied and 3D vision-language models +1
HATS: High-Accuracy Triple-Set Watermarking for Large Language ModelsLLM safety and security → Watermarking generated content and models
Learning General Policies with Policy Gradient MethodsDeep reinforcement learning methods → Deep RL algorithms and representations
Interpretable Hybrid Deep Q-Learning Framework for IoT-Based Food Spoilage Prediction with Synthetic Data Generation and Hardware ValidationDeep reinforcement learning methods → RL for energy, networks and operations
Generative vector search to improve pathology foundation models across multimodal vision-language tasksRetrieval-augmented generation → Medical and biomedical RAG
First-Order Representation Languages for Goal-Conditioned RLDeep reinforcement learning methods → Deep RL algorithms and representations
ReasonCD: A Multimodal Reasoning Large Model for Implicit Change-of-Interest Semantic MiningLLM reasoning → Spatial and embodied reasoning
PENDULUM: A Benchmark for Assessing Sycophancy in Multimodal Large Language ModelsMultimodal LLMs / vision-language models → VLM benchmarks and visual reasoning
OMP: One-step Meanflow Policy with Directional AlignmentDiffusion models → Diffusion and flow policies for robotics
SafeMed-R1: Adversarial Reinforcement Learning for Generalizable and Robust Medical Reasoning in Vision-Language ModelsLLM reasoning → Clinical and biomedical applications +1
Bridging Semantics and Geometry: A Decoupled LVLM-SAM Framework for Reasoning Segmentation in Optical Remote SensingLLM reasoning → Spatial and embodied reasoning
RMLer: Synthesizing Novel Objects across Diverse Categories via Reinforcement Mixing LearningDeep reinforcement learning methods → RL with diffusion and flow models
Helios: A Foundational Language Model for Smart Energy Knowledge Reasoning and ApplicationLLM reasoning → Knowledge- and tool-augmented reasoning
Vibe Reasoning: Eliciting Frontier AI Mathematical Capabilities -- A Case Study on IMO 2025 Problem 6LLM agents and tool use → General LLM agent frameworks and benchmarks
Are All Data Necessary? Efficient Data Pruning for Large-scale Autonomous Driving Dataset via Trajectory Entropy MaximizationEnd-to-end autonomous driving → Imitation-learning end-to-end planners
Machine Unlearning in the Era of Quantum Machine Learning: An Empirical StudyLLM safety and security → Machine unlearning
Auto-Prompting with Retrieval Guidance for Frame Detection in LogisticsLLM reasoning → Prompting and chain-of-thought reasoning
ChemATP: A Training-Free Chemical Reasoning Framework for Large Language ModelsLLM reasoning → Clinical and biomedical applications
DeliveryBench: Can Agents Earn Profit in Real World?LLM agents and tool use → General LLM agent frameworks and benchmarks
Selective LoRA for Visual Tokens and Attention HeadsMultimodal LLMs / vision-language models → Prompt learning and adaptation of CLIP
Semantically-Equivalent Transformations-Based Backdoor Attacks against Neural Code Models: Characterization and MitigationLLM safety and security → Backdoor and data poisoning attacks
MixKVQ: Query-Aware Mixed-Precision KV Cache Quantization for Long-Context ReasoningLLM reasoning → Test-time compute and efficient thinking +1
CascadeInfer: Length-Aware Scheduling of LLM Serving with Low Latency and Load BalancingEfficient LLM inference and serving → Serving systems and request scheduling
JEPA-Reasoner: Decoupling Latent Reasoning from Token GenerationLLM reasoning → Test-time compute and efficient thinking
A Flexible Field-Based Policy Learning Framework for Diverse Robotic Systems and SensorsDiffusion models → Diffusion and flow policies for robotics
Understanding Chain-of-Thought in Large Language Models via Topological Data AnalysisLLM reasoning → Prompting and chain-of-thought reasoning
QuCo-RAG: Quantifying Uncertainty from the Pre-training Corpus for Dynamic Retrieval-Augmented GenerationRetrieval-augmented generation → RAG robustness and adaptive retrieval
WorldRFT: Latent World Model Planning with Reinforcement Fine-Tuning for Autonomous DrivingWorld models → World models for autonomous driving +1
AWPO: Enhancing Tool-Use of Large Language Models through Adaptive Integration of Reasoning RewardsLLM reasoning → Reinforcement learning for reasoning
Stop saying LLM: Large Discourse Models (LDM) and Artificial Discursive Agent (ADA)?LLM agents and tool use → Social simulation and multi-agent debate
Generative Giants, Retrieval Weaklings: Why do Multimodal Large Language Models Fail at Multimodal Retrieval?Multimodal LLMs / vision-language models → Document understanding and multimodal RAG +1
FC-MIR: A Mobile Screen Awareness Framework for Intent-Aware Recommendation based on Frame-Compressed Multimodal Trajectory ReasoningLLM reasoning → Multimodal reasoning
Explicit and Non-asymptotic Query Complexities of Rank-Based Zeroth-order Algorithm on Stochastic Smooth FunctionsLLM alignment / RLHF / preference optimisation → Reward modeling and RLHF
Mamba-Based Modality Disentanglement Network for Multi-Contrast MRI ReconstructionState-space sequence models (S4, Mamba) → Mamba for medical image analysis
Tool-Augmented Hybrid Ensemble Reasoning with Distillation for Bilingual Mathematical Problem SolvingLLM reasoning → Mathematical reasoning +1
A Large Language Model Based Method for Complex Logical Reasoning over Knowledge GraphsLLM reasoning → Knowledge- and tool-augmented reasoning
γ(3,4) `Attention' in Cognitive Agents: Ontology-Free Knowledge Representations With Promise Theoretic SemanticsLLM agents and tool use → Web, search, and research agents
Population-Evolve: a Parallel Sampling and Evolutionary Method for LLM Math ReasoningLLM reasoning → Mathematical reasoning
Watch Closely: Mitigating Object Hallucinations in Large Vision-Language Models with Disentangled DecodingMultimodal LLMs / vision-language models → Hallucination mitigation in LVLMs
WaTeRFlow: Watermark Temporal Robustness via Flow ConsistencyDiffusion models → Robustness, privacy and adversarial aspects
The Erasure Illusion: Stress-Testing the Generalization of LLM Forgetting EvaluationLLM safety and security → Machine unlearning
Steering Vision-Language Pre-trained Models for Incremental Face Presentation Attack DetectionMultimodal LLMs / vision-language models → MLLM jailbreaks and adversarial robustness
DREAM: Dynamic Red-teaming across Environments for AI ModelsLLM agents and tool use → Agent threats, guardrails and governance +1
Do You Really Need a GPU to Guard Your LLM? CPU-Class Classifiers and Multi-Stage Pipelines for Safety Enforcement at ScaleLLM safety and security → Safety alignment, refusal and guardrails
Context-Aware Initialization for Reducing Generative Path Length in Diffusion Language ModelsDiffusion models → Discrete and text diffusion
Enhancing Fully Formatted End-to-End Speech Recognition with Knowledge Distillation via Multi-Codebook Vector QuantizationKnowledge distillation → Speech and audio distillation
Lag Operator SSMs: A Geometric Framework for Structured State Space ModelingState-space sequence models (S4, Mamba) → Core SSM architectures for language
Online Robust Reinforcement Learning with General Function ApproximationDeep reinforcement learning methods → RL theory and sample complexity
Training Multimodal Large Reasoning Models Needs Better Thoughts: A Three-Stage Framework for Long Chain-of-Thought Synthesis and SelectionLLM reasoning → Multimodal reasoning

All 324