Papers

New on arXiv

Mon 26 Jan 2026

368 papers

HyEm: Query-Adaptive Hyperbolic Retrieval for Biomedical Ontologies via Euclidean Vector IndexingRetrieval-augmented generation → Medical and biomedical RAG
Beyond Offline A/B Testing: Context-Aware Agent Simulation for Recommender System EvaluationLLM agents and tool use → Agentic recommender systems
The Truth, the Whole Truth, and Nothing but the Truth: Automatic Visualization Evaluation from Reconstruction QualityLLM agents and tool use → Multimodal and embodied agents
Exploring the Use of VLMs for Navigation Assistance for People with Blindness and Low VisionLLM reasoning → Spatial and embodied reasoning +1
ReasonNavi: Human-Inspired Global Map Reasoning for Zero-Shot Embodied NavigationLLM reasoning → Spatial and embodied reasoning
Not the Example, but the Process: How Self-Generated Examples Enhance LLM ReasoningLLM reasoning → Reasoning benchmarks and evaluation +1
Enhancing Action and Ingredient Modeling for Semantically Grounded Recipe GenerationMultimodal LLMs / vision-language models → Image editing and quality assessment
Reranker Optimization via Geodesic Distances on k-NN ManifoldsRetrieval-augmented generation → Generation and context integration
From Transcripts to AI Agents: Knowledge Extraction, RAG Integration, and Robust Evaluation of Conversational AI AssistantsLLM agents and tool use → General LLM agent frameworks and benchmarks
Computability of Agentic SystemsLLM agents and tool use → General LLM agent frameworks and benchmarks
SITUATE -- Synthetic Object Counting Dataset for VLM trainingMultimodal LLMs / vision-language models → VLM benchmarks and visual reasoning
LTS-VoiceAgent: A Listen-Think-Speak Framework for Efficient Streaming Voice Interaction via Semantic Triggering and Incremental ReasoningLLM reasoning → Audio-language model reasoning
A Unifying View of Coverage in Linear Off-Policy EvaluationDeep reinforcement learning methods → RL theory and sample complexity
Randomization Boosts KV Caching, Learning Balances Query Load: A Joint PerspectiveEfficient LLM inference and serving → KV cache optimization
FreeOrbit4D: Training-Free Arbitrary Camera Redirection for Monocular Videos via Foreground-Complete 4D ReconstructionDiffusion models → 3D, point cloud and scene generation
LLMs versus the Halting Problem: Characterizing Program Termination ReasoningLLM reasoning → Reasoning for agents, planning and code
Save the Good Prefix: Precise Error Penalization via Process-Supervised RL to Enhance LLM ReasoningLLM reasoning → Reinforcement learning for reasoning
Reinforcement Learning for Quantum TechnologyDeep reinforcement learning methods → Quantum reinforcement learning
Vector-Valued Distributional Reinforcement Learning Policy Evaluation: A Hilbert Space Embedding ApproachDeep reinforcement learning methods → RL theory and sample complexity
A Few Bad Neurons: Isolating and Surgically Correcting SycophancyMechanistic interpretability → Activation steering methods
BabyReasoningBench: Generating Developmentally-Inspired Reasoning Tasks for Evaluating Baby Language ModelsLLM reasoning → Reasoning benchmarks and evaluation
Advances in Diffusion-Based Generative CompressionDiffusion models → Diffusion for compression and wireless
How Is Uncertainty Propagated in Knowledge Distillation?Knowledge distillation → General knowledge distillation methods
Implicit Q-Learning and SARSA: Liberating Policy Control from Step-Size CalibrationDeep reinforcement learning methods → RL theory and sample complexity
Flatter Tokens are More Valuable for Speculative Draft Model TrainingEfficient LLM inference and serving → Speculative decoding
XProvence: Zero-Cost Multilingual Context Pruning for Retrieval-Augmented GenerationRetrieval-augmented generation → Generation and context integration
MulVul: Retrieval-augmented Multi-Agent Code Vulnerability Detection via Cross-Model Prompt EvolutionRetrieval-augmented generation → Privacy and security of RAG +1
Reuse your FLOPs: Scaling RL on Hard Problems by Conditioning on Very Off-Policy PrefixesLLM reasoning → Reinforcement learning for reasoning
MortalMATH: Evaluating the Conflict Between Reasoning Objectives and Emergency ContextsLLM reasoning → Reasoning for agents, planning and code
Preference-Conditioned Multi-Objective Reinforcement Learning for Tactical Decision Making in Heavy-Duty VehiclesDeep reinforcement learning methods → RL for autonomous driving and traffic
POPE: Learning to Reason on Hard Problems via Privileged On-Policy ExplorationLLM reasoning → Reinforcement learning for reasoning
Teaching Models to Teach Themselves: Reasoning at the Edge of LearnabilityLLM reasoning → Reinforcement learning for reasoning
PRECISE: Reducing the Bias of LLM Evaluations Using Prediction-Powered Ranking EstimationRetrieval-augmented generation → Generation and context integration
Dep-Search: Learning Dependency-Aware Reasoning Traces with Persistent MemoryLLM reasoning → Knowledge- and tool-augmented reasoning
Beyond Preferences: Learning Alignment Principles Grounded in Human Reasons and ValuesLLM reasoning → Reasoning benchmarks and evaluation +1
α3-SecBench: A Large-Scale Evaluation Suite of Security, Resilience, and Trust for LLM-based UAV Agents over 6G NetworksLLM agents and tool use → Multimodal and embodied agents
HalluGuard: Demystifying Data-Driven and Reasoning-Driven Hallucinations in LLMsLLM reasoning → Hallucination detection and mitigation
Trust, Don't Trust, or Flip: Robust Preference-Based Reinforcement Learning with Multi-Expert FeedbackLLM alignment / RLHF / preference optimisation → Reward modeling and RLHF +1
Let's Make Every Pull Request Meaningful: An Empirical Analysis of Developer and Agentic Pull RequestsLLM agents and tool use → Software engineering coding agents
The P-Completeness of Inverted Index Traversal: On the Complexity of Evaluating Boolean Query DAGsLLM reasoning → Knowledge- and tool-augmented reasoning
TSRBench: A Comprehensive Multi-task Multi-modal Time Series Reasoning Benchmark for Generalist ModelsLLM reasoning → Time series reasoning
Why Keep Your Doubts to Yourself? Trading Visual Uncertainties in Multi-Agent Bandit SystemsMultimodal LLMs / vision-language models → Embodied and 3D vision-language models +1
Self-Distilled Reasoner: On-Policy Self-Distillation for Large Language ModelsLLM reasoning → Prompting and chain-of-thought reasoning +1
Advances and Innovations in the Multi-Agent Robotic System (MARS) ChallengeMultimodal LLMs / vision-language models → Embodied and 3D vision-language models +1
One Adapts to Any: Meta Reward Modeling for Personalized LLM AlignmentLLM alignment / RLHF / preference optimisation → Human values and AI feedback alignment
Reflect: Transparent Principle-Guided Reasoning for Constitutional Alignment at ScaleLLM reasoning → Reinforcement learning for reasoning +1
Gained in Translation: Privileged Pairwise Judges Enhance Multilingual ReasoningLLM reasoning → Reasoning benchmarks and evaluation
From Fuzzy to Exact: The Halo Architecture for Infinite-Depth Reasoning via Rational ArithmeticLLM reasoning → Mathematical reasoning
TEA-Bench: A Systematic Benchmarking of Tool-enhanced Emotional Support Dialogue AgentLLM agents and tool use → Social simulation and multi-agent debate
Mechanistic Analysis of Catastrophic Forgetting in Large Language Models During Continual Fine-tuningMixture-of-experts → Mixture of LoRA experts
A Pragmatic VLA Foundation ModelVision-language-action models → Generalist VLA models and benchmarks
Learned harmonic mean estimation of the marginal likelihood for multimodal posteriors with flow matchingDiffusion models → Flow matching, time series and science
ART for Diffusion Sampling: A Reinforcement Learning Approach to Timestep ScheduleDiffusion models → Diffusion and flow policies for robotics +1
S2GR: Stepwise Semantic-Guided Reasoning in Latent Space for Generative RecommendationLLM reasoning → LLM reasoning for recommendation
FaLW: A Forgetting-aware Loss Reweighting for Long-tailed UnlearningLLM safety and security → Machine unlearning
FadeMem: Biologically-Inspired Forgetting for Efficient Agent MemoryLLM agents and tool use → Agent memory systems
AdaReasoner: Dynamic Tool Orchestration for Iterative Visual ReasoningLLM reasoning → Multimodal reasoning +2
Rank-1 Approximation of Inverse Fisher for Natural Policy Gradients in Deep Reinforcement LearningDeep reinforcement learning methods → RL theory and sample complexity
Adaptive Domain Shift in Diffusion Models for Cross-Modality Image TranslationDiffusion models → Image generation, text-to-image and editing
CASSANDRA: Programmatic and Probabilistic Learning and Inference for Stochastic World ModelingWorld models → World models for language agents
Geometry-Free Conditional Diffusion Modeling for Solving the Inverse Electrocardiography ProblemDiffusion models → Medical imaging and segmentation
A Balanced Neuro-Symbolic Approach for Commonsense Abductive LogicLLM reasoning → Reasoning benchmarks and evaluation
Learning long term climate-resilient transport adaptation pathways under direct and indirect flood impacts using reinforcement learningDeep reinforcement learning methods → RL for energy, networks and operations
GimmBO: Interactive Generative Image Model Merging via Bayesian OptimizationDiffusion models → Diffusion foundations and sampling
From Classification to Ranking: Enhancing LLM Reasoning Capabilities for MBTI Personality DetectionLLM reasoning → Reasoning benchmarks and evaluation
K-Myriad: Jump-starting reinforcement learning with unsupervised parallel agentsDeep reinforcement learning methods → RL theory and sample complexity
GenAgent: Scaling Text-to-Image Generation via Agentic Multimodal ReasoningLLM reasoning → Multimodal reasoning +1
From Verifiable Dot to Reward Chain: Harnessing Verifiable Reference-based Rewards for Reinforcement Learning of Open-ended GenerationLLM alignment / RLHF / preference optimisation → Process reward models for reasoning +1
Exploring Fine-Tuning for In-Context Retrieval and Efficient KV-Caching in Long-Context Language ModelsRetrieval-augmented generation → Generation and context integration +1
Just-In-Time Reinforcement Learning: Continual Learning in LLM Agents Without Gradient UpdatesLLM agents and tool use → Agent memory systems
DEEPMED: Building a Medical DeepResearch Agent via Multi-hop Med-Search Data and Turn-Controlled Agentic Training & InferenceLLM reasoning → Clinical and biomedical applications +1
AgentDoG: A Diagnostic Guardrail Framework for AI Agent Safety and SecurityLLM agents and tool use → Agent threats, guardrails and governance +1
OffSeeker: Online Reinforcement Learning Is Not All You Need for Deep Research AgentsLLM agents and tool use → Web, search, and research agents +1
Geneses: Unified Generative Speech Enhancement and SeparationDiffusion models → Audio, speech and music
3DGesPolicy: Phoneme-Aware Holistic Co-Speech Gesture Generation Based on Action ControlDiffusion models → Audio, speech and music
SG-CADVLM: A Context-Aware Decoding Powered Vision Language Model for Safety-Critical Scenario GenerationMultimodal LLMs / vision-language models → Driving VLMs and open-vocabulary detection
daVinci-Dev: Agent-native Mid-training for Software EngineeringLLM agents and tool use → Software engineering coding agents
Do not be greedy, Think Twice: Sampling and Selection for Document-level Information ExtractionLLM reasoning → Knowledge- and tool-augmented reasoning
ARMOR: Agentic Reasoning for Methods Orchestration and Reparameterization for Robust Adversarial AttacksLLM reasoning → Reasoning for agents, planning and code +1
Dynamic Thinking-Token Selection for Efficient Reasoning in Large Reasoning ModelsLLM reasoning → Test-time compute and efficient thinking
AI Agent for Reverse-Engineering Legacy Finite-Difference Code and Translating to DevitoLLM agents and tool use → Web, search, and research agents
Making medical vision-language models think causally across modalities with retrieval-augmented cross-modal reasoningLLM reasoning → Clinical and biomedical applications +2
Code over Words: Overcoming Semantic Inertia via Code-Grounded ReasoningLLM reasoning → Reasoning benchmarks and evaluation
Adapter Merging Reactivates Latent Reasoning Traces: A Mechanism AnalysisLLM reasoning → Test-time compute and efficient thinking
Q-Bench-Portrait: Benchmarking Multimodal Large Language Models on Portrait Image Quality PerceptionMultimodal LLMs / vision-language models → MLLM training and visual instruction tuning
Promises, Perils, and (Timely) Heuristics for Mining Coding Agent ActivityLLM agents and tool use → Software engineering coding agents
Agentic Much? Adoption of Coding Agents on GitHubLLM agents and tool use → Software engineering coding agents
TC-IDM: Grounding Video Generation for Executable Zero-shot Robot MotionWorld models → Embodied and robot world models +2
Integrating Fine-Grained Audio-Visual Evidence for Robust Multimodal Emotion ReasoningLLM reasoning → Audio-language model reasoning
MultiVis-Agent: A Multi-Agent Framework with Logic Rules for Reliable and Comprehensive Cross-Modal Data VisualizationLLM agents and tool use → Multimodal and embodied agents
SwipeGen: Bridging the Execution Gap in GUI Agents via Human-like Swipe SynthesisLLM agents and tool use → GUI and mobile agents
Temp-R1: A Unified Autonomous Agent for Complex Temporal KGQA via Reverse Curriculum Reinforcement LearningDeep reinforcement learning methods → RL training of language and agent models
Reinforcement Learning with Distributed MPC for Fuel-Efficient Platoon Control with Discrete Gear TransitionsDeep reinforcement learning methods → RL for autonomous driving and traffic
TriPlay-RL: Tri-Role Self-Play Reinforcement Learning for LLM Safety AlignmentLLM alignment / RLHF / preference optimisation → Safety alignment and jailbreak attacks +2
U-Fold: Dynamic Intent-Aware Context Folding for User-Centric AgentsLLM agents and tool use → General LLM agent frameworks and benchmarks
Think-Augmented Function Calling: Improving LLM Parameter Accuracy Through Embedded ReasoningLLM reasoning → Reasoning for agents, planning and code +1
When Nobody Around Is Real: Exploring Public Opinions and User Experiences On the Multi-Agent AI Social PlatformLLM agents and tool use → Social simulation and multi-agent debate
Orchestrating Specialized Agents for Trustworthy Enterprise RAGRetrieval-augmented generation → Agentic RAG, agents and memory
BoRP: Bootstrapped Regression Probing for Scalable and Human-Aligned LLM EvaluationSynthetic data and self-improvement for LLMs → LLM synthetic data generation
V-Loop: Visual Logical Loop Verification for Hallucination Detection in Medical Visual Question AnsweringMultimodal LLMs / vision-language models → Hallucination mitigation in LVLMs

All 368