Papers

New on arXiv

Fri 20 Feb 2026

245 papers

ACE-TA: An Agentic Teaching Assistant for Grounded Q&A, Quiz Generation, and Code TutoringLLM agents and tool use → Educational agents and math problem solving
Tuning Qwen2.5-VL to Improve Its Web Interaction SkillsMultimodal LLMs / vision-language models → Embodied and 3D vision-language models
EvoDiagram: Agentic Editable Diagram Creation via Design Expertise EvolutionLLM agents and tool use → Multimodal and embodied agents
Agentic AI, Retrieval-Augmented Generation, and the Institutional Turn: Legal Architectures and Financial Governance in the Age of Distributional AGILLM agents and tool use → Financial trading and forecasting agents
Think First, Diffuse Fast: Improving Diffusion Language Model Reasoning via Autoregressive Plan ConditioningLLM reasoning → Diffusion reasoning and test-time scaling +1
Reward Under Attack: Analyzing the Robustness and Hackability of Process Reward ModelsLLM reasoning → Reinforcement learning for reasoning +1
GraphSkill: Documentation-Guided Hierarchical Retrieval-Augmented Coding for Complex Graph ReasoningLLM reasoning → Knowledge- and tool-augmented reasoning +1
Not all tokens are needed(NAT): token efficient reinforcement learningLLM reasoning → Reinforcement learning for reasoning +1
Distilling and Adapting: A Topology-Aware Framework for Zero-Shot Interaction Prediction in Multiplex Biological NetworksKnowledge distillation → Graph neural network distillation
Annealed Co-Generation: Disentangling Variables via Progressive Pairwise ModelingDiffusion models → Diffusion foundations and sampling
Correlation Analysis of Generative ModelsDiffusion models → Flow matching, time series and science
Multimodal Modular Chain of Thoughts in Energy Performance Certificate AssessmentLLM reasoning → Multimodal reasoning
NutriOrion: A Hierarchical Multi-Agent Framework for Personalized Nutrition Intervention Grounded in Clinical GuidelinesLLM agents and tool use → Medical and scientific domain agents
Adaptive Time Series Reasoning via Segment SelectionLLM reasoning → Time series reasoning
Modeling and Recovering Hierarchical Structural Architectures of ROS 2 Systems from Code and Launch Configurations using LLM-based AgentsLLM agents and tool use → Software engineering coding agents
Decoding ML Decision: An Agentic Reasoning Framework for Large-Scale Ranking SystemLLM agents and tool use → Web, search, and research agents
Learning Invariant Visual Representations for Planning with Joint-Embedding Predictive World ModelsWorld models → Latent predictive world models
Effect of Patch Size on Fine-Tuning Vision Transformers in Two-Dimensional and Three-Dimensional Medical Image ClassificationMultimodal LLMs / vision-language models → Medical vision-language models
MapTab: A Diagnostic Benchmark for Long-Horizon Multi-Criteria Multimodal Reasoning on Heterogeneous Topological GraphsLLM reasoning → Spatial and embodied reasoning +1
DM4CT: Benchmarking Diffusion Models for Computed Tomography ReconstructionDiffusion models → Medical imaging and segmentation
Luna-2: Scalable Single-Token Evaluation with Small Language ModelsLLM safety and security → Safety alignment, refusal and guardrails
Hierarchical Reward Design from Language: Enhancing Alignment of Agent Behavior with Human SpecificationsLLM alignment / RLHF / preference optimisation → Reward modeling and RLHF
Debug2Fix: Can Interactive Debugging Help Coding Agents Fix More Bugs?LLM agents and tool use → Software engineering coding agents
RPU -- A Reasoning Processing UnitLLM reasoning → Test-time compute and efficient thinking
Rodent-BenchMultimodal LLMs / vision-language models → Embodied and 3D vision-language models
Morphological Addressing of Identity Basins in Text-to-Image Diffusion ModelsDiffusion models → Image generation, text-to-image and editing
VLANeXt: Recipes for Building Strong VLA ModelsMultimodal LLMs / vision-language models → Embodied and 3D vision-language models +1
Deep Reinforcement Learning for Optimizing Energy Consumption in Smart Grid SystemsDeep reinforcement learning methods → RL for energy, networks and operations
JAEGER: Joint 3D Audio-Visual Grounding and Reasoning in Simulated Physical EnvironmentsLLM reasoning → Audio-language model reasoning
Going Down Memory Lane: Scaling Tokens for Video Stream Understanding with Dynamic KV-Cache MemoryEfficient LLM inference and serving → Visual token pruning in multimodal models
SARAH: Spatially Aware Real-time Agentic HumansLLM agents and tool use → Multimodal and embodied agents
VIRAASAT: Traversing Novel Paths for Indian Cultural ReasoningLLM reasoning → Reasoning benchmarks and evaluation
The Geometry of Noise: Why Diffusion Models Don't Need Noise ConditioningDiffusion models → Diffusion foundations and sampling
CapNav: Benchmarking Vision Language Models on Capability-conditioned Indoor NavigationMultimodal LLMs / vision-language models → Embodied and 3D vision-language models
Generated Reality: Human-centric World Simulation using Interactive Video Generation with Hand and Camera ControlWorld models → Embodied and robot world models +1
Exploiting Completeness Perception with Diffusion Transformer for Unified 3D MRI SynthesisDiffusion models → Medical imaging and segmentation
How Fast Can I Run My VLA? Demystifying VLA Inference Performance with VLA-PerfVision-language-action models → Efficient VLA inference and adaptation
Rethinking Flow and Diffusion Bridge Models for Speech EnhancementDiffusion models → Audio, speech and music
Learning Smooth Time-Varying Linear Policies with an Action Jacobian PenaltyDeep reinforcement learning methods → Deep RL algorithms and representations
Analyzing and Improving Chain-of-Thought Monitorability Through Information TheoryLLM reasoning → Prompting and chain-of-thought reasoning
Diffusing to Coordinate: Efficient Online Multi-Agent Diffusion PoliciesDiffusion models → Diffusion and flow policies for robotics
PRISM: Parallel Reward Integration with Symmetry for MORLDeep reinforcement learning methods → RL for games and combinatorial optimization
Simplifying Outcomes of Language Model Component Analyses with ELIAMechanistic interpretability → Knowledge and representations inside LLMs
Thinking by Subtraction: Confidence-Driven Contrastive Decoding for LLM ReasoningLLM reasoning → Test-time compute and efficient thinking
SimVLA: A Simple VLA Baseline for Robotic ManipulationVision-language-action models → Generalist VLA models and benchmarks
RAT+: Train Dense, Infer Sparse -- Recurrence Augmented Attention for Dilated InferenceEfficient LLM inference and serving → KV cache optimization
BLM-Guard: Explainable Multimodal Ad Moderation with Chain-of-Thought and Policy-Aligned RewardsLLM reasoning → Multimodal reasoning
Improving Sampling for Masked Diffusion Models via Information GainDiffusion models → Discrete and text diffusion
Can AI Lower the Barrier to Cybersecurity? A Human-Centered Mixed-Methods Study of Novice CTF LearningLLM agents and tool use → Agent threats, guardrails and governance
FENCE: A Financial and Multimodal Jailbreak Detection DatasetMultimodal LLMs / vision-language models → MLLM jailbreaks and adversarial robustness +1
Agentic Adversarial QA for Improving Domain-Specific LLMsLLM agents and tool use → General LLM agent frameworks and benchmarks
Flow Matching with Injected Noise for Offline-to-Online Reinforcement LearningDiffusion models → Diffusion and flow policies for robotics +1
TempoNet: Slack-Quantized Transformer-Guided Reinforcement Scheduler for Adaptive Deadline-Centric Real-Time DispatchsDeep reinforcement learning methods → RL for energy, networks and operations
MeanVoiceFlow: One-step Nonparallel Voice Conversion with Mean FlowsDiffusion models → Audio, speech and music
Interacting safely with cyclists using Hamilton-Jacobi reachability and reinforcement learningDeep reinforcement learning methods → RL for autonomous driving and traffic
Neurosymbolic Language Reasoning as Satisfiability Modulo TheoryLLM reasoning → Reasoning benchmarks and evaluation
OODBench: Out-of-Distribution Benchmark for Large Vision-Language ModelsMultimodal LLMs / vision-language models → VLM benchmarks and visual reasoning
Predict to Skip: Linear Multistep Feature Forecasting for Efficient Diffusion TransformersDiffusion models → Diffusion foundations and sampling
Balancing Symmetry and Efficiency in Graph Flow MatchingDiffusion models → Flow matching, time series and science
Continual-NExT: A Unified Comprehension And Generation Continual Learning FrameworkMultimodal LLMs / vision-language models → Prompt learning and adaptation of CLIP
Gradient Regularization Mitigates Reward Hacking in Reinforcement Learning from Human Feedback and Verifiable RewardsLLM alignment / RLHF / preference optimisation → Reward modeling and RLHF
Mean-Field Reinforcement Learning without SynchronyDeep reinforcement learning methods → RL for games and combinatorial optimization
Cross-Embodiment Offline Reinforcement Learning for Heterogeneous Robot DatasetsOffline RL and imitation learning from data → Real-world robot offline RL
Dual-Channel Attention Guidance for Training-Free Image Editing Control in Diffusion TransformersDiffusion models → Image generation, text-to-image and editing
UAOR: Uncertainty-aware Observation Reinjection for Vision-Language-Action ModelsMultimodal LLMs / vision-language models → Embodied and 3D vision-language models +1
DeepSVU: Towards In-depth Security-oriented Video Understanding via Unified Physical-world Regularized MoEMixture-of-experts → Vision mixture-of-experts
Flow Actor-Critic for Offline Reinforcement LearningOffline RL and imitation learning from data → Offline RL algorithms
Are LLMs Ready for Neural-integrated Mechanistic Modeling? A Benchmark and Agentic FrameworkLLM agents and tool use → General LLM agent frameworks and benchmarks
Policy Gradient Algorithms in Average-Reward Multichain MDPsDeep reinforcement learning methods → RL theory and sample complexity
Aurora: Neuro-Symbolic AI Driven Advising AgentLLM agents and tool use → Educational agents and math problem solving
Turbo Connection: Reasoning as Information Flow from Higher to Lower LayersLLM reasoning → Prompting and chain-of-thought reasoning
WorkflowPerturb: Calibrated Stress Tests for Evaluating Multi-Agent Workflow MetricsLLM agents and tool use → Software engineering coding agents
Decomposing Retrieval Failures in RAG for Long-Document Financial Question AnsweringRetrieval-augmented generation → Legal and financial RAG
Learning Optimal and Sample-Efficient Decision Policies with GuaranteesDeep reinforcement learning methods → RL theory and sample complexity
ROCKET: Residual-Oriented Multi-Layer Alignment for Spatially-Aware Vision-Language-Action ModelsVision-language-action models → 3D and spatial-aware VLA
Analyzing LLM Instruction Optimization for Tabular Fact VerificationLLM reasoning → Knowledge- and tool-augmented reasoning
MIRA: Memory-Integrated Reinforcement Learning Agent with Limited LLM GuidanceLLM agents and tool use → Agent memory systems
Alignment in Time: Peak-Aware Orchestration for Long-Horizon Agentic SystemsLLM agents and tool use → General LLM agent frameworks and benchmarks
LLM Spirals of Delusion: A Benchmarking Audit Study of AI Chatbot Interfaces
FIGURA: A Modular Prompt Engineering Method for Artistic Figure Photography in Safety-Filtered Text-to-Image Models
Do Large Language Models Possess a Theory of Mind? A Comparative Evaluation Using the Strange Stories Paradigm
A foundation model for electrodermal activity data
Multi-Axis Trust Modeling for Interpretable Account Hijacking Detection
Automating Document Intelligence in Statutory City Planning
Automating the Analysis and Improvement of Dynamic Programming Algorithms with Applications to Natural Language Processing
Self-hosted Lecture-to-Quiz: Local LLM MCQ Generation with Deterministic Quality Control
Evo: Autoregressive-Diffusion Large Language Models with Evolving Balance
RACER: Risk-Aware Calibrated Efficient Routing for Large Language Models
OptiRoulette Optimizer: A New Stochastic Meta-Optimizer for up to 5.3x Faster Convergence
Consensus is Not Verification: Why Crowd Wisdom Strategies Fail for LLM Truthfulness
A Novel Approach for Testing Water Safety Using Deep Learning Inference of Microscopic Images of Unincubated Water Samples
A systematic approach to answering the easy problems of consciousness based on an executable cognitive system
CogGen: Cognitive-Load-Inspired Fully Unsupervised Deep Generative Modeling for Compressively Sampled MRI Reconstruction
Descriptor: Parasitoid Wasps and Associated Hymenoptera Dataset (DAPWH)
Toward AI Autonomous Navigation for Mechanical Thrombectomy using Hierarchical Modular Multi-agent Reinforcement Learning (HM-MARL)
Large Causal Models for Temporal Causal Discovery
Communication-Efficient Personalized Adaptation via Federated-Local Model Merging
PolyFrame at MWE-2026 AdMIRe 2: When Words Are Not Enough: Multimodal Idiom Disambiguation
Global Low-Rank, Local Full-Rank: The Holographic Encoding of Learned Algorithms
Noise Scheduling as Information-Guided Allocation in Diffusion Training

All 245