Papers

New on arXiv

Fri 27 Feb 2026

392 papers

Steerable Instruction Following Coding Data Synthesis with Actor-Parametric Schema Co-EvolutionSynthetic data and self-improvement for LLMs → Code and instruction data synthesis
TableNet A Large-Scale Table Dataset with LLM-Powered AutonomousLLM reasoning → Knowledge- and tool-augmented reasoning
LSR: Linguistic Safety Robustness Benchmark for Low-Resource West African LanguagesLLM alignment / RLHF / preference optimisation → Safety alignment and jailbreak attacks +1
From Tokens To Agents: A Researcher's Guide To Understanding Large Language ModelsLLM agents and tool use → General LLM agent frameworks and benchmarks
Do Diffusion Models Dream of Electric Planes? Discrete and Continuous Simulation-Based Inference for Aircraft DesignDiffusion models → Flow matching, time series and science
ICaRus: Identical Cache Reuse for Efficient Multi Model InferenceLLM agents and tool use → Software engineering coding agents +1
Demand Acceptance using Reinforcement Learning for Dynamic Vehicle Routing Problem with Emission QuotaDeep reinforcement learning methods → RL for games and combinatorial optimization
Learning Retrieval Models with Sparse AutoencodersMechanistic interpretability → SAE features and their applications
PREBA: Surgical Duration Prediction via PCA-Weighted Retrieval-Augmented LLMs and Bayesian Averaging AggregationRetrieval-augmented generation → Medical and biomedical RAG
Learning from Partial Chain-of-Thought via Truncated-Reasoning Self-DistillationLLM reasoning → Prompting and chain-of-thought reasoning +1
Architecture-Aware LLM Inference Optimization on AMD Instinct GPUs: A Comprehensive Benchmark and Deployment StudyMixture-of-experts → Efficient MoE training and inference systems
ObjChangeVR: Object State Change Reasoning from Continuous Egocentric Views in VR EnvironmentsLLM reasoning → Spatial and embodied reasoning +1
Omni-C: Compressing Heterogeneous Modalities into a Single Dense EncoderMixture-of-experts → Multimodal MoE models
Induced Numerical Instability: Hidden Costs in Multimodal Large Language ModelsMultimodal LLMs / vision-language models → MLLM training and visual instruction tuning
CamDirector: Towards Long-Term Coherent Video Trajectory EditingDiffusion models → Video generation
DiffSOS: Acoustic Conditional Diffusion Model for Speed-of-Sound Reconstruction in Ultrasound Computed TomographyDiffusion models → Medical imaging and segmentation
Verifier-Bound Communication for LLM Agents: Certified Bounds on Covert SignalingLLM agents and tool use → Agent threats, guardrails and governance
NeuroHex: A Brain-Inspired Hex Coordinate System to Enable Highly Computationally-Efficient World Models for Continuous Online-Adaptive LearningWorld models → World models for navigation
COOP2: Defining, Observing, and Repairing Cooperation in LLM Multi-Agent SystemsLLM agents and tool use → General LLM agent frameworks and benchmarks
Proof-of-Perception: Certified Tool-Using Multimodal Reasoning with Compositional Conformal GuaranteesLLM reasoning → Multimodal reasoning
How Well Do Multimodal Models Reason on ECG Signals?LLM reasoning → Clinical and biomedical applications
DIG to Heal: Scaling General-purpose Agent Collaboration via Explainable Dynamic Decision PathsLLM agents and tool use → General LLM agent frameworks and benchmarks
When does Chain-of-Thought Help: A Markovian PerspectiveLLM reasoning → Prompting and chain-of-thought reasoning
Stepwise Penalization for Length-Efficient Chain-of-Thought ReasoningLLM reasoning → Test-time compute and efficient thinking
TraderBench: How Robust Are AI Agents in Adversarial Capital Markets?LLM agents and tool use → Financial trading and forecasting agents
Multi-Sourced, Multi-Agent Evidence Retrieval for Fact-CheckingLLM agents and tool use → Web, search, and research agents
Agentic Scientific Simulation: Execution-Grounded Model Construction and ReconstructionLLM agents and tool use → Web, search, and research agents
VisRef: Visual Refocusing while Thinking Improves Test-Time Scaling in Multi-Modal Large Reasoning ModelsLLM reasoning → Multimodal reasoning
TACIT Benchmark: A Programmatic Visual Reasoning Benchmark for Generative and Discriminative ModelsLLM reasoning → Multimodal reasoning
Efficient Flow Matching for Sparse-View CT ReconstructionDiffusion models → Medical imaging and segmentation
Stateful Token Reduction for Long-Video Hybrid VLMsState-space sequence models (S4, Mamba) → Vision Mamba backbones +1
Formal Analysis and Supply Chain Security for Agentic AI SkillsLLM agents and tool use → Agent threats, guardrails and governance
SKeDA: A Generative Watermarking Framework for Text-to-video Diffusion ModelsDiffusion models → Robustness, privacy and adversarial aspects +1
ST-Lite: Training-Free KV Cache Compression with Spatio-Trajectory Guidance for Long-Horizon GUI AgentsEfficient LLM inference and serving → Visual token pruning in multimodal models
ClarEval: A Benchmark for Evaluating Clarification Skills of Code Agents under Ambiguous InstructionsLLM agents and tool use → Software engineering coding agents
Mode Seeking meets Mean Seeking for Fast Long Video GenerationDiffusion models → Video generation
CUDA Agent: Large-Scale Agentic RL for High-Performance CUDA Kernel GenerationLLM agents and tool use → Software engineering coding agents
Taming Momentum: Rethinking Optimizer States Through Low-Rank ApproximationLLM reasoning → Test-time compute and efficient thinking
Resources for Automated Evaluation of Assistive RAG Systems that Help Readers with News Trustworthiness AssessmentRetrieval-augmented generation → RAG pipeline design and evaluation
A Minimal Agent for Automated Theorem ProvingLLM agents and tool use → Educational agents and math problem solving
Enhancing Spatial Understanding in Image Generation via Reward ModelingLLM alignment / RLHF / preference optimisation → Diffusion and multimodal model alignment
From Leaky Thoughts to Private Reasoning: Controlling What LRMs Say to ThemselvesLLM reasoning → Reasoning for agents, planning and code
SenCache: Accelerating Diffusion Model Inference via Sensitivity-Aware CachingDiffusion models → Diffusion foundations and sampling
Uncertainty Quantification for Multimodal Large Language Models with Incoherence-adjusted Semantic VolumeMultimodal LLMs / vision-language models → MLLM training and visual instruction tuning
Multi-Objective Reinforcement Learning for Large-Scale Tote Allocation in Human-Robot Collaborative Fulfillment CentersDeep reinforcement learning methods → RL for energy, networks and operations
ArgLLM-App: An Interactive System for Argumentative Reasoning with Large Language ModelsLLM reasoning → Reasoning for agents, planning and code
HumanOrbit: 3D Human Reconstruction as 360deg Orbit GenerationDiffusion models → 3D, point cloud and scene generation
Robust Skills, Brittle Grounding: Diagnosing Restricted Generalization in Vision-Language Action Policies via Multi-Object PickingVision-language-action models → Failure recovery and robustness
CoME: Empowering Channel-of-Mobile-Experts with Informative Hybrid-Capabilities ReasoningLLM agents and tool use → GUI and mobile agents
AgenticOCR: Parsing Only What You Need for Efficient Retrieval-Augmented GenerationRetrieval-augmented generation → Multimodal RAG
Online World Modeling Enables Real-World Inverse Reinforcement Learning from ObservationWorld models → Embodied and robot world models +2
Agentic AI-RAN: Enabling Intent-Driven, Explainable and Self-Evolving Open RAN IntelligenceLLM agents and tool use → Agentic AI for wireless networks
Toward Guarantees for Clinical Reasoning in Vision Language Models via Formal VerificationLLM reasoning → Clinical and biomedical applications +1
Recycling Failures: Salvaging Exploration in RLVR via Fine-Grained Off-Policy GuidanceLLM reasoning → Reinforcement learning for reasoning
Artificial Agency Program: Curiosity, compression, and communication in agentsDeep reinforcement learning methods → Deep RL algorithms and representations
Preference Packing: Efficient Preference Optimization for Large Language ModelsLLM alignment / RLHF / preference optimisation → LLM preference optimization methods
Adaptive Correlation-Weighted Intrinsic Rewards for Reinforcement LearningDeep reinforcement learning methods → Deep RL algorithms and representations
A Novel Hierarchical Multi-Agent System for Payments Using LLMsLLM agents and tool use → General LLM agent frameworks and benchmarks
Task Complexity Matters: An Empirical Study of Reasoning in LLMs for Sentiment AnalysisLLM reasoning → Audio-language model reasoning
Quant Experts: Token-aware Adaptive Error Reconstruction with Mixture of Experts for Large Vision-Language Models QuantizationMultimodal LLMs / vision-language models → Visual token pruning and efficiency +1
Look Carefully: Adaptive Visual Reinforcements in Multimodal Large Language Models for Hallucination MitigationMultimodal LLMs / vision-language models → Hallucination mitigation in LVLMs
RewardUQ: A Unified Framework for Uncertainty-Aware Reward ModelsLLM alignment / RLHF / preference optimisation → Reward modeling and RLHF
Portfolio Reinforcement Learning with Scenario-Context RolloutDeep reinforcement learning methods → Adversarial robustness and RL in finance
Curriculum Reinforcement Learning for Quadrotor Racing with Random ObstaclesDeep reinforcement learning methods → RL for robot navigation and UAV control
GuardAlign: Test-time Safety Alignment in Multimodal Large Language ModelsMultimodal LLMs / vision-language models → MLLM jailbreaks and adversarial robustness +2
Steering and Rectifying Latent Representation Manifolds in Frozen Multi-modal LLMs for Video Anomaly DetectionMultimodal LLMs / vision-language models → Chart understanding and anomaly detection
Interpretable Debiasing of Vision-Language Models for Social FairnessLLM reasoning → Reasoning benchmarks and evaluation +1
Ordinal Diffusion Models for Color Fundus ImagesDiffusion models → Medical imaging and segmentation
Jailbreak Foundry: From Papers to Runnable Attacks for Reproducible BenchmarkingLLM safety and security → Jailbreaking
Foundation World Models for Agents that Learn, Verify, and Adapt Reliably Beyond Static EnvironmentsWorld models → World models for language agents
Venus: Benchmarking and Empowering Multimodal Large Language Models for Aesthetic Guidance and CroppingMultimodal LLMs / vision-language models → Image editing and quality assessment
Pessimistic Auxiliary Policy for Offline Reinforcement LearningOffline RL and imitation learning from data → Offline RL algorithms
Learning Generation Orders for Masked Discrete Diffusion Models via Variational InferenceDiffusion models → Discrete and text diffusion
Thinking with Images as Continuous Actions: Numerical Visual Chain-of-ThoughtLLM reasoning → Multimodal reasoning +1
SwitchCraft: Training-Free Multi-Event Video Generation with Attention ControlsDiffusion models → Video generation
HotelQuEST: Balancing Quality and Efficiency in Agentic SearchLLM agents and tool use → Web, search, and research agents
PointCoT: A Multi-modal Benchmark for Explicit 3D Geometric ReasoningLLM reasoning → Spatial and embodied reasoning +1
MemEmo: Evaluating Emotion in Memory Systems of AgentsLLM agents and tool use → Agent memory systems
Learning to Build: Autonomous Robotic Assembly of Stable Structures Without Predefined PlansDeep reinforcement learning methods → RL for robot locomotion and manipulation
Ref-Adv: Exploring MLLM Visual Reasoning in Referring Expression TasksLLM reasoning → Multimodal reasoning +1
TSC: Topology-Conditioned Stackelberg Coordination for Multi-Agent Reinforcement Learning in Interactive DrivingEnd-to-end autonomous driving → Cooperative and V2X driving
DACESR: Degradation-Aware Conditional Embedding for Real-World Image Super-ResolutionState-space sequence models (S4, Mamba) → Mamba for image restoration
LK Losses: Direct Acceptance Rate Optimization for Speculative DecodingEfficient LLM inference and serving → Speculative decoding
RF-Agent: Automated Reward Function Design via Language Agent Tree SearchLLM agents and tool use → Reinforcement learning for LLM agents
Bandwidth-adaptive Cloud-Assisted 360-Degree 3D Perception for Autonomous VehiclesEnd-to-end autonomous driving → VLMs for autonomous driving
SWE-rebench V2: Language-Agnostic SWE Task Collection at ScaleLLM agents and tool use → Software engineering coding agents
RUMAD: Reinforcement-Unifying Multi-Agent DebateLLM reasoning → Reasoning for agents, planning and code +1
Polarization Uncertainty-Guided Diffusion Model for Color Polarization Image DemosaickingDiffusion models → Inverse problems and image restoration
Footprint-Guided Exemplar-Free Continual Histopathology Report GenerationMultimodal LLMs / vision-language models → Medical vision-language models
Learning to maintain safety through expert demonstrations in settings with unknown constraints: A Q-learning perspectiveDeep reinforcement learning methods → Safe and constrained RL
Beyond State-Wise Mirror Descent: Offline Policy Optimization with Parametric PoliciesOffline RL and imitation learning from data → Offline RL algorithms +1
See, Act, Adapt: Active Perception for Unsupervised Cross-Domain Visual Adaptation via Personalized VLM-Guided AgentLLM agents and tool use → Multimodal and embodied agents
Actor-Critic Pretraining for Proximal Policy OptimizationOffline RL and imitation learning from data → Imitation learning methods +1
EMO-R3: Reflective Reinforcement Learning for Emotional Reasoning in Multimodal Large Language ModelsLLM reasoning → Audio-language model reasoning +1
MPU: Towards Secure and Privacy-Preserving Knowledge Unlearning for Large Language ModelsLLM safety and security → Machine unlearning
Divide and Conquer: Accelerating Diffusion-Based Large Language Models via Adaptive Parallel DecodingEfficient LLM inference and serving → Diffusion LLM and MoE inference
Diffusion Probe: Generated Image Result Prediction Using CNN ProbesDiffusion models → Image generation, text-to-image and editing
Reasoning-Driven Multimodal LLM for Domain GeneralizationLLM reasoning → Multimodal reasoning +1
MAGE: Multi-scale Autoregressive Generation for Offline Reinforcement LearningOffline RL and imitation learning from data → Offline RL algorithms
OPTIAGENT: A Physics-Driven Agentic Framework for Automated Optical DesignLLM agents and tool use → Multimodal and embodied agents

All 392