Papers

New on arXiv

Mon 16 Feb 2026

316 papers

Beyond the Prompt in Large Language Models: Comprehension, In-Context Learning, and Chain-of-ThoughtLLM reasoning → Prompting and chain-of-thought reasoning
A Retrieval-Augmented Language Assistant for Unmanned Aircraft Safety Assessment and Regulatory ComplianceRetrieval-augmented generation → Legal and financial RAG
Building the ethical AI framework of the future: from philosophy to practiceLLM agents and tool use → Agent threats, guardrails and governance
Thin Keys, Full Values: Reducing KV Cache via Low-Dimensional Attention SelectionEfficient LLM inference and serving → KV cache optimization
Generating Realistic, Protocol-Compliant Maritime Radio Dialogues using Self-Instruct and Low-Rank AdaptationSynthetic data and self-improvement for LLMs → Synthetic reasoning data and bootstrapping
Engineering Reasoning and Instruction (ERI) Benchmark: A Large Taxonomy-driven Dataset for Foundation Models and AgentsLLM reasoning → Reasoning benchmarks and evaluation +1
Is Mamba Reliable for Medical Imaging?State-space sequence models (S4, Mamba) → Mamba for medical image analysis
Fast and Effective On-policy Distillation from Reasoning PrefixesKnowledge distillation → On-policy and self-distillation for LLMs
How to Train Your Long-Context Visual Document ModelMultimodal LLMs / vision-language models → Video understanding with MLLMs
MyoInteract: A Framework for Fast Prototyping of Biomechanical HCI Tasks using Reinforcement LearningDeep reinforcement learning methods → RL for robot locomotion and manipulation
Ground-Truth Depth in Vision Language Models: Spatial Context Understanding in Conversational AI for XR-Robotic Support in Emergency First ResponseLLM reasoning → Spatial and embodied reasoning +1
Automatically Finding Reward Model BiasesLLM alignment / RLHF / preference optimisation → Reward modeling and RLHF
Multi-Agent Home Energy Management AssistantLLM agents and tool use → Financial trading and forecasting agents
Secure and Energy-Efficient Wireless Agentic AI NetworksLLM agents and tool use → Agentic AI for wireless networks
Colosseum: Auditing Collusion in Cooperative Multi-Agent SystemsLLM agents and tool use → Financial trading and forecasting agents
OpaqueToolsBench: Learning Nuances of Tool Behavior Through InteractionLLM agents and tool use → Tool learning and function calling
AIC CTU@AVerImaTeC: dual-retriever RAG for image-text fact checkingRetrieval-augmented generation → Multimodal RAG
Seeing to Generalize: How Visual Data Corrects Binding ShortcutsMultimodal LLMs / vision-language models → VLM benchmarks and visual reasoning
Mind the (DH) Gap! A Contrast in Risky Choices Between Reasoning and Conversational LLMsLLM reasoning → Reasoning benchmarks and evaluation +1
Protecting Language Models Against Unauthorized Distillation through Trace RewritingLLM safety and security → Watermarking generated content and models +1
Zero-shot HOI Detection with MLLM-based Detector-agnostic Interaction RecognitionMultimodal LLMs / vision-language models → Driving VLMs and open-vocabulary detection
ResearchGym: Evaluating Language Model Agents on Real-World AI ResearchLLM agents and tool use → Web, search, and research agents
Mixture-of-Experts under Finite-Rate Gating: Communication--Generalization Trade-offsMixture-of-experts → Expert routing strategies
The Agentic Automation Canvas: a structured framework for agentic AI project designLLM agents and tool use → General LLM agent frameworks and benchmarks
Near-Optimal Sample Complexity for Online Constrained MDPsDeep reinforcement learning methods → RL theory and sample complexity
Image Generation with a Sphere EncoderDiffusion models → Diffusion foundations and sampling
Symmetry in language statistics shapes the geometry of model representationsMechanistic interpretability → Sparse autoencoder training and evaluation
Rethinking Diffusion Models with Symmetries through Canonicalization with Applications to Molecular Graph GenerationDiffusion models → Flow matching, time series and science
Hunt Globally: Wide Search AI Agents for Drug Asset Scouting in Investing, Business Development, and Competitive IntelligenceLLM agents and tool use → Medical and scientific domain agents
Scaling Beyond Masked Diffusion Language ModelsDiffusion models → Discrete and text diffusion
Cold-Start Personalization via Training-Free Priors from Structured World ModelsWorld models → World models for language agents
BPP: Long-Context Robot Imitation Learning by Focusing on Key History FramesOffline RL and imitation learning from data → Robot manipulation imitation learning
Efficient Sampling with Discrete Diffusion Models: Sharp and Adaptive GuaranteesDiffusion models → Diffusion foundations and sampling
Learning User Interests via Reasoning and Distillation for Cross-Domain News RecommendationLLM reasoning → LLM reasoning for recommendation
Boundary Point Jailbreaking of Black-Box LLMsLLM safety and security → Jailbreaking
ThermEval: A Structured Benchmark for Evaluation of Vision-Language Models on Thermal ImageryMultimodal LLMs / vision-language models → VLM benchmarks and visual reasoning
DM0: An Embodied-Native Vision-Language-Action Model towards Physical AIVision-language-action models → Generalist VLA models and benchmarks
Counterfactual Fairness Evaluation of LLM-Based Contact Center Agent Quality Assurance SystemLLM agents and tool use → General LLM agent frameworks and benchmarks
PhyScensis: Physics-Augmented LLM Agents for Complex Physical Scene ArrangementLLM agents and tool use → Multimodal and embodied agents
Sovereign Agents: Towards Infrastructural Sovereignty and Diffused Accountability in Decentralized AILLM agents and tool use → Agent threats, guardrails and governance
Kami of the Commons: Towards Designing Agentic AI to Steward the CommonsLLM agents and tool use → General LLM agent frameworks and benchmarks
Variance-Reduced (varepsilon,δ)-Unlearning using Forget Set GradientsLLM safety and security → Machine unlearning
MAC-AMP: A Closed-Loop Multi-Agent Collaboration System for Multi-Objective Antimicrobial Peptide DesignLLM agents and tool use → Medical and scientific domain agents
ReusStdFlow: A Standardized Reusability Framework for Dynamic Workflow Construction in Agentic AILLM agents and tool use → Software engineering coding agents
BFS-PO: Best-First Search for Large Reasoning ModelsLLM reasoning → Test-time compute and efficient thinking
The Potential of CoT for Reasoning: A Closer Look at Trace DynamicsLLM reasoning → Prompting and chain-of-thought reasoning
Picking the Right Specialist: Attentive Neural Process-based Selection of Task-Specialized Models as Tools for Agentic Healthcare SystemsLLM agents and tool use → Medical and scientific domain agents
Lifted Relational Probabilistic Inference via Implicit LearningLLM reasoning → Reasoning benchmarks and evaluation
Model Context Protocol (MCP) Tool Descriptions Are Smelly! Towards Improving AI Agent Efficiency with Augmented MCP Tool DescriptionsLLM agents and tool use → Tool learning and function calling
Affordance Transfer Across Object Instances via Semantically Anchored Functional MapOffline RL and imitation learning from data → Robot manipulation imitation learning
On the Emergence of Implicit Curriculum in RLVR Learning DynamicsLLM reasoning → Reinforcement learning for reasoning
Concept Influence: Leveraging Interpretability to Improve Performance and Efficiency in Training Data AttributionMechanistic interpretability → Knowledge and representations inside LLMs
Goldilocks RL: Tuning Task Difficulty to Escape Sparse Rewards for ReasoningLLM reasoning → Reinforcement learning for reasoning
EmbeWebAgent: Embedding Web Agents into Any Customized UILLM agents and tool use → Web, search, and research agents
World Models for Policy Refinement in StarCraft IIWorld models → World models for language agents +1
Interactionless Inverse Reinforcement Learning: A Data-Centric Framework for Durable AlignmentOffline RL and imitation learning from data → Inverse reinforcement learning
RF-GPT: Teaching AI to See the Wireless WorldMultimodal LLMs / vision-language models → MLLM training and visual instruction tuning
Physical Commonsense Reasoning for Lower-Resourced Languages and Dialects: a Study on BasqueLLM reasoning → Reasoning benchmarks and evaluation
Overthinking Loops in Agents: A Structural Risk via MCP ToolsLLM reasoning → Reasoning for agents, planning and code +1
A Geometric Analysis of Small-sized Language Model HallucinationsLLM agents and tool use → Medical and scientific domain agents
Emergently Misaligned Language Models Show Behavioral Self-Awareness That Shifts With Subsequent RealignmentLLM alignment / RLHF / preference optimisation → LLM preference optimization methods
Multi-Agent Comedy Club: Investigating Community Discussion Effects on LLM Humor GenerationLLM agents and tool use → Social simulation and multi-agent debate
Unlocking Reasoning Capability on Machine Translation in Large Language ModelsLLM reasoning → Reasoning benchmarks and evaluation
Rethinking the Role of LLMs in Time Series ForecastingLLM reasoning → Time series reasoning
AI Arms and Influence: Frontier Models Exhibit Sophisticated Reasoning in Simulated Nuclear CrisesLLM reasoning → Reasoning for agents, planning and code
ManeuverNet: A Soft Actor-Critic Framework for Precise Maneuvering of Double-Ackermann-Steering Robots with Optimized Reward FunctionsDeep reinforcement learning methods → RL for robot locomotion and manipulation
WebWorld: A Large-Scale World Model for Web Agent TrainingWorld models → World models for language agents +1
Evolutionary System Prompt Learning for Reinforcement Learning in LLMsLLM agents and tool use → General LLM agent frameworks and benchmarks
Harness Engineering for Agentic AI Coding Tools: An Exploratory StudyLLM agents and tool use → Software engineering coding agents
Exposing the Systematic Vulnerability of Open-Weight Models to Prefill AttacksLLM safety and security → Jailbreaking
SynthSAEBench: Evaluating Sparse Autoencoders on Scalable Realistic Synthetic DataSynthetic data and self-improvement for LLMs → LLM synthetic data generation +1
ST-EVO: Towards Generative Spatio-Temporal Evolution of Multi-Agent Communication TopologiesLLM agents and tool use → General LLM agent frameworks and benchmarks
Universal Image Immunization against Diffusion-based Image Editing via Semantic InjectionDiffusion models → Robustness, privacy and adversarial aspects
GradMAP: Faster Layer Pruning with Gradient Metric and Projection CompensationLLM reasoning → Test-time compute and efficient thinking
VIGIL: Tackling Hallucination Detection in Image RecontextualizationMultimodal LLMs / vision-language models → Hallucination mitigation in LVLMs
Towards Selection as Power: Bounding Decision Authority in Autonomous AgentsLLM agents and tool use → Agent threats, guardrails and governance
MATEO: A Multimodal Benchmark for Temporal Reasoning and Planning in LVLMsLLM reasoning → Spatial and embodied reasoning
Decoupled Continuous-Time Reinforcement Learning via Hamiltonian FlowDeep reinforcement learning methods → RL theory and sample complexity
RNM-TD3: N:M Semi-structured Sparse Reinforcement Learning From ScratchDeep reinforcement learning methods → Deep RL algorithms and representations
DriveFine: Refining-Augmented Masked Diffusion VLA for Precise and Robust DrivingVision-language-action models → VLA for autonomous driving +4
Simulation-based Learning of Electrical Cabinet Assembly Using Robot SkillsDeep reinforcement learning methods → RL for robot locomotion and manipulation
Fluid-Agent Reinforcement LearningDeep reinforcement learning methods → RL for games and combinatorial optimization
Governing AI Forgetting: Auditing for Machine Unlearning ComplianceLLM safety and security → Machine unlearning
Replanning Human-Robot Collaborative Tasks with Vision-Language Models via Semantic and Physical Dual-CorrectionMultimodal LLMs / vision-language models → Embodied and 3D vision-language models
MoRL: Reinforced Reasoning for Unified Motion Understanding and GenerationLLM reasoning → Spatial and embodied reasoning
Diagnosing Knowledge Conflict in Multimodal Long-Chain ReasoningLLM reasoning → Reasoning benchmarks and evaluation +1
Large Language Models for Math Education in Low-Resource Languages: A Study in Sinhala and TamilLLM reasoning → Mathematical reasoning
Efficient Text-Guided Convolutional Adapter for the Diffusion ModelDiffusion models → Image generation, text-to-image and editing
Formally Verifying and Explaining Sepsis Treatment Policies with COOL-MCDeep reinforcement learning methods → RL for clinical treatment and surgery
TikArt: Stabilizing Aperture-Guided Fine-Grained Visual Reasoning with Reinforcement LearningLLM reasoning → Multimodal reasoning +1
Learning Transferability: A Two-Stage Reinforcement Learning Approach for Enhancing Quadruped Robots' Performance in U-Shaped Stair ClimbingDeep reinforcement learning methods → RL for robot locomotion and manipulation
Socially-Weighted Alignment: A Game-Theoretic Framework for Multi-Agent LLM SystemsLLM agents and tool use → Social simulation and multi-agent debate
HyperRAG: Reasoning N-ary Facts over Hypergraphs for Retrieval Augmented GenerationLLM reasoning → Knowledge- and tool-augmented reasoning +1
Measuring and Mitigating Post-hoc Rationalization in Reverse Chain-of-Thought GenerationLLM reasoning → Prompting and chain-of-thought reasoning
CoCoDiff: Correspondence-Consistent Diffusion Model for Fine-grained Style TransferDiffusion models → Image generation, text-to-image and editing
Traceable Latent Variable Discovery Based on Multi-Agent CollaborationLLM agents and tool use → Medical and scientific domain agents
WiSparse: Boosting LLM Inference Efficiency with Weight-Aware Mixed Activation SparsityEfficient LLM inference and serving → Early exit and edge inference
Precedent-Informed Reasoning: Mitigating Overthinking in Large Reasoning Models via Test-Time Precedent LearningLLM reasoning → Test-time compute and efficient thinking
Broken Chains: The Cost of Incomplete Reasoning in LLMsLLM reasoning → Test-time compute and efficient thinking
RoboSolver: A Multi-Agent Large Language Model Framework for Solving Robotic Arm ProblemsLLM agents and tool use → Multimodal and embodied agents

All 316