Papers

New on arXiv

Wed 24 Dec 2025

220 papers

LLM-Driven Preference Data Synthesis for Proactive Prediction of the Next User Utterance in Human-Machine DialogueSynthetic data and self-improvement for LLMs → Code and instruction data synthesis
From Values to Frameworks: A Qualitative Study of Ethical Reasoning in Agentic AI PractitionersLLM agents and tool use → General LLM agent frameworks and benchmarks
Governing Cloud Data Pipelines with Agentic AILLM agents and tool use → Software engineering coding agents
Logic Sketch Prompting (LSP): A Deterministic and Interpretable Prompting MethodLLM reasoning → Prompting and chain-of-thought reasoning
Agentic Software Issue Resolution with Large Language Models: A SurveyLLM agents and tool use → Software engineering coding agents
Shape of Thought: When Distribution Matters More than Correctness in Reasoning TasksLLM reasoning → Prompting and chain-of-thought reasoning
dUltra: Ultra-Fast Diffusion Language Models via Reinforcement LearningDiffusion models → Discrete and text diffusion
Cerberus: Multi-Agent Reasoning and Coverage-Guided Exploration for Static Detection of Runtime ErrorsLLM agents and tool use → Software engineering coding agents
EVE: A Generator-Verifier System for Generative PoliciesDiffusion models → Diffusion and flow policies for robotics
A Tool Bottleneck Framework for Clinically-Informed and Interpretable Medical Image UnderstandingMultimodal LLMs / vision-language models → Medical vision-language models
A Survey of Freshness-Aware Wireless Networking with Reinforcement LearningDeep reinforcement learning methods → RL for energy, networks and operations
Understanding Virality: A Rubric based Vision-Language Model Framework for Short-Form Edutainment EvaluationLLM reasoning → Multimodal reasoning +1
A Reinforcement Learning Approach to Synthetic Data GenerationLLM safety and security → Membership inference attacks +1
AInsteinBench: Benchmarking Coding Agents on Scientific RepositoriesLLM agents and tool use → Web, search, and research agents
A Graph-Augmented knowledge Distillation based Dual-Stream Vision Transformer with Region-Aware Attention for Gastrointestinal Disease Classification with Explainable AIKnowledge distillation → Distillation for medical imaging
HiStream: Efficient High-Resolution Video Generation via Redundancy-Eliminated StreamingDiffusion models → Video generation
Beyond Memorization: A Multi-Modal Ordinal Regression Benchmark to Expose Popularity Bias in Vision-Language ModelsMultimodal LLMs / vision-language models → VLM benchmarks and visual reasoning
Optimizing Decoding Paths in Masked Diffusion Models by Quantifying UncertaintyDiffusion models → Discrete and text diffusion
Fast SAM2 with Text-Driven Token PruningEfficient LLM inference and serving → Token pruning and merging for ViTs
Your Reasoning Benchmark May Not Test Reasoning: Revealing Perception Bottleneck in Abstract Reasoning BenchmarksLLM reasoning → Multimodal reasoning
A Plan Reuse Mechanism for LLM-Driven AgentLLM agents and tool use → General LLM agent frameworks and benchmarks
AndroidLens: Long-latency Evaluation with Nested Sub-targets for Android GUI AgentsLLM agents and tool use → GUI and mobile agents
Model Merging via Multi-Teacher Knowledge DistillationKnowledge distillation → General knowledge distillation methods
SMART SLM: Structured Memory and Reasoning Transformer, A Small Language Model for Accurate Document AssistanceLLM reasoning → Reasoning benchmarks and evaluation
ACD: Direct Conditional Control for Video Diffusion Models via Attention SupervisionDiffusion models → Video generation
ReaSeq: Unleashing World Knowledge via Reasoning for Sequential ModelingLLM reasoning → LLM reasoning for recommendation
CoTDeceptor:Adversarial Code Obfuscation Against CoT-Enhanced LLM Code AgentsLLM agents and tool use → Agent threats, guardrails and governance
Casting a SPELL: Sentence Pairing Exploration for LLM Limitation-breakingLLM safety and security → Jailbreaking
RoboCade: Gamifying Robot Data CollectionOffline RL and imitation learning from data → Robot manipulation imitation learning
MiST: Understanding the Role of Mid-Stage Scientific Training in Developing Chemical Reasoning ModelsLLM reasoning → Clinical and biomedical applications
RoboSafe: Safeguarding Embodied Agents via Executable Safety LogicLLM agents and tool use → Agent threats, guardrails and governance
Latent Implicit Visual ReasoningLLM reasoning → Multimodal reasoning +1
VisRes Bench: On Evaluating the Visual Reasoning Capabilities of VLMsLLM reasoning → Multimodal reasoning +1
AutoBaxBuilder: Bootstrapping Code Security BenchmarkingSynthetic data and self-improvement for LLMs → Code and instruction data synthesis
Active inference and artificial reasoningWorld models → World models for language agents
MarineEval: Assessing the Marine Intelligence of Vision-Language ModelsMultimodal LLMs / vision-language models → Embodied and 3D vision-language models
STLDM: Spatio-Temporal Latent Diffusion Model for Precipitation NowcastingDiffusion models → Flow matching, time series and science
Beyond Context: Large Language Models' Failure to Grasp Users' IntentLLM safety and security → Safety alignment, refusal and guardrails
FreeInpaint: Tuning-free Prompt Alignment and Visual Rationality Enhancement in Image InpaintingDiffusion models → 3D, point cloud and scene generation
Dyna-Style Reinforcement Learning Modeling and Control of Non-linear DynamicsDeep reinforcement learning methods → RL theory and sample complexity
Agentic Explainable Artificial Intelligence (Agentic XAI) Approach To Explore Better Explanation: A Case Study in Decision Support for Rice Cultivation in JapanLLM agents and tool use → Multimodal and embodied agents
Multi-Attribute guided Thermal Face Image Translation based on Latent Diffusion ModelDiffusion models → Image generation, text-to-image and editing
Policy-Conditioned Policies for Multi-Agent Task SolvingLLM agents and tool use → Reinforcement learning for LLM agents
Enhancing diffusion models with Gaussianization preprocessingDiffusion models → Diffusion foundations and sampling
Rethinking Supervised Fine-Tuning: Emphasizing Key Answer Tokens for Improved LLM AccuracyLLM reasoning → Test-time compute and efficient thinking
FluencyVE: Marrying Temporal-Aware Mamba with Bypass Attention for Video EditingDiffusion models → Video generation +1
GateBreaker: Gate-Guided Attacks on Mixture-of-Expert LLMsLLM safety and security → Mixture-of-experts routing safety
Learning from Next-Frame Prediction: Autoregressive Video Modeling Encodes Effective RepresentationsDiffusion models → Video generation
Distilling the Essence: Efficient Reasoning Distillation via Sequence TruncationLLM reasoning → Prompting and chain-of-thought reasoning +1
TrafficSimAgent: A Hierarchical Agent Framework for Autonomous Traffic Simulation with MCP ControlLLM agents and tool use → Multimodal and embodied agents
PUFM++: Point Cloud Upsampling via Enhanced Flow MatchingDiffusion models → 3D, point cloud and scene generation
AegisAgent: An Autonomous Defense Agent Against Prompt Injection Attacks in LLM-HARsLLM agents and tool use → Agent threats, guardrails and governance +1
A Blockchain-Monitored Agentic AI Architecture for Trusted Perception-Reasoning-Action PipelinesLLM agents and tool use → Agent threats, guardrails and governance
X-ray Insights Unleashed: Pioneering the Enhancement of Multi-Label Long-Tail DataDiffusion models → Medical imaging and segmentation
Meta-RL with Bayesian Linear Task ModelsLLM alignment / RLHF / preference optimisation → Reward modeling and RLHF +1
DAO-Agent: Zero Knowledge-Verified Incentives for Decentralized Multi-Agent CoordinationLLM agents and tool use → Agent threats, guardrails and governance
Generalization of Diffusion Models Arises with a Balanced Representation SpaceDiffusion models → Diffusion foundations and sampling
Can Agentic AI Match the Performance of Human Data Scientists?LLM agents and tool use → Web, search, and research agents
One Tool Is Enough: Reinforcement Learning for Repository-Level LLM AgentsLLM agents and tool use → Software engineering coding agents
Reflection Pretraining Enables Token-Level Self-Correction in Biological Sequence ModelsLLM reasoning → Clinical and biomedical applications
Pioneering Multimodal Emotion Recognition in the Era of Large Models: From Closed Sets to Open VocabulariesMultimodal LLMs / vision-language models → Multimodal emotion understanding
Reasoning-Driven Amodal Completion: Collaborative Agents and Perceptual EvaluationLLM reasoning → Spatial and embodied reasoning +1
Transductive Visual Programming: Evolving Tool Libraries from Experience for Spatial ReasoningLLM reasoning → Spatial and embodied reasoning +1
Self-supervised Multiplex Consensus Mamba for General Image FusionState-space sequence models (S4, Mamba) → Point clouds, detection and remote sensing
RevFFN: Memory-Efficient Full-Parameter Fine-Tuning of Mixture-of-Experts LLMs with Reversible BlocksMixture-of-experts → MoE large language models
MMSRARec: Summarization and Retrieval Augumented Sequential Recommendation Based on Multimodal Large Language ModelMultimodal LLMs / vision-language models → Document understanding and multimodal RAG +1
Model-free stochastic linear quadratic control for discrete-time systems with multiplicative and additive noises via semidefinite programmingDeep reinforcement learning methods → RL for games and combinatorial optimization
Where Did This Sentence Come From? Tracing Provenance in LLM Reasoning DistillationLLM reasoning → Prompting and chain-of-thought reasoning +1
DiEC: Diffusion Embedded ClusteringDiffusion models → Diffusion foundations and sampling
Benchmarking and Enhancing VLM for Compressed Image UnderstandingMultimodal LLMs / vision-language models → Visual token pruning and efficiency
The Silent Scholar Problem: A Probabilistic Framework for Breaking Epistemic Asymmetry in LLM AgentsRetrieval-augmented generation → Agentic RAG, agents and memory +1
Proprioception Enhances Vision Language Model in Generating Captions and Subtask Segmentations for Robot TaskMultimodal LLMs / vision-language models → Embodied and 3D vision-language models
Robustness Certificates for Neural Networks Against Data Poisoning and Evasion AttacksLLM safety and security → Backdoor and data poisoning attacks
Opportunities and Challenges of Natural Language Processing for Low-Resource Senegalese Languages in Social Science Research
Introducing Axlerod: An LLM-based Chatbot for Assisting Independent Insurance Agents
Evaluating Novelty in AI-Generated Research Plans Using Multi-Workflow LLM Pipelines
The Environmental Impact of AI Servers and Sustainable Solutions
AgentTutor: Empowering Personalized Learning with Multi-Turn Interactive Teaching in Intelligent Education Systems
LLM_annotate: A Python package for annotating and analyzing fiction characters
Tree of Preferences for Diversified Recommendation
Decomposing LLM Self-Correction: The Accuracy-Correction Paradox and Error Depth Hypothesis
Speak the Art: A Direct Speech to Image Generation Framework
A multi-algorithm approach for operational human resources workload balancing in a last mile urban delivery system
LLMTM: Benchmarking and Optimizing LLMs for Temporal Motif Analysis in Dynamic Graphs
LuxIA: A Lightweight Unitary matriX-based Framework Built on an Iterative Algorithm for Photonic Neural Network Training
Evaluating an Adaptive Multispectral Turret System for Autonomous Tracking Across Variable Illumination Conditions
INSIGHT: Spatially resolved survival modelling from routine histology crosslinked with molecular profiling reveals prognostic epithelial-immune axes in stage II/III colorectal cancer
The Physics Constraint Paradox: When Removing Explicit Constraints Improves Physics-Informed Data for Machine Learning
ReVEAL: GNN-Guided Reverse Engineering for Formal Verification of Optimized Multipliers
Cardiac mortality prediction in patients undergoing PCI based on real and synthetic data
Analyzing Skill Element in Online Fantasy Cricket
Graph Attention-based Adaptive Transfer Learning for Link Prediction
Interpretable Perturbation Modeling Through Biomedical Knowledge Graphs
Temporal Visual Semantics-Induced Human Motion Understanding with Large Language Models
Amortized Inference for Model Rocket Aerodynamics: Learning to Estimate Physical Parameters from Simulation
The Affine Divergence: Aligning Activation Updates Beyond Normalisation
An Equivariance Toolbox for Learning Dynamics
Fuzzwise: Intelligent Initial Corpus Generation for Fuzzing
Morality is Contextual: Learning Interpretable Moral Contexts from Human Data with Probabilistic Clustering and Large Language Models
Dynamic Attention (DynAttn): Interpretable High-Dimensional Spatio-Temporal Forecasting (with Application to Conflict Fatalities)

All 220