Papers

New on arXiv

Mon 13 Jul 2026

429 papers

Think Before You Comfort: Reflective Cognitive Alignment for Protocol-Grounded Elderly Stimulation AgentsLLM reasoning → Clinical and biomedical applications +1
Improving LLM Interpretability with User-Centric Chain-of-Thought ReasoningLLM reasoning → Reasoning benchmarks and evaluation
IDP AutoOpt: Agent-Driven Optimization of Document Processing Pipeline ConfigurationsLLM agents and tool use → Web, search, and research agents
IFCMemoryBench: Evaluating Long-Term Memory of LLM-Based Agents in BIM Information RetrievalLLM agents and tool use → Agent memory systems
RRS-10K: A Multitask Vision-Language Model Benchmark for Rare Remote Sensing Image InterpretationMultimodal LLMs / vision-language models → Embodied and 3D vision-language models
SevDiff: Severity-Conditioned Diffusion for Long-Tail Conflict Trajectory GenerationDiffusion models → Diffusion and flow policies for robotics
StrideDiffusion: Accelerating Diffusion Models for Time-series GenerationDiffusion models → Diffusion foundations and sampling
FedDP-PALD: A Privacy-Preserving Federated Latent Diffusion Framework with Prototype Aggregation for Medical Data SynthesisDiffusion models → Robustness, privacy and adversarial aspects
A2BM: Alignment-Aware Bridge Matching for Image-to-Image TranslationDiffusion models → Flow matching, time series and science
Self-Improving AI Coding Agents Through Accumulated Behavioral Rules: A Closed-Loop FrameworkLLM agents and tool use → Software engineering coding agents +1
Securing LLMs in the Wild: Privacy and Security Challenges at the EdgeLLM safety and security → Membership inference attacks
Baselines Before Architecture: Evaluating Coding Agents for Autonomous Penetration TestingLLM agents and tool use → Agent threats, guardrails and governance
Phantom Guardrails: When Self-Improving Agent Harnesses Fix Failures That Never HappenedLLM agents and tool use → Software engineering coding agents +2
Beyond AI-Generated Labels: Watermarking, Co-Creation, and Conflation of AI-Generation with DisinformationLLM safety and security → Watermarking generated content and models
SingGuard-NSFA: Extensible Guardrails for Agentic AI via Generative Reasoning and Real-Time ClassificationLLM agents and tool use → Agent threats, guardrails and governance +1
Inference Economics of Enterprise Coding Agents: A Case Study of Cloud vs. On-Premise LLMsLLM agents and tool use → Agent memory systems
Contract-Grounded Behavior Tree Synthesis via Coding AgentsLLM agents and tool use → General LLM agent frameworks and benchmarks
Fine-Tuned Multi-Agent Framework for Detecting OCEAN in Life NarrativesLLM agents and tool use → Social simulation and multi-agent debate
Cost-Governed RAG: Unified Per-Tenant Cost Attribution Across Retrieval and Generation in Multi-Tenant LLM SystemsRetrieval-augmented generation → RAG pipeline design and evaluation
Self-Consistent Flow: Unifying Velocity and Endpoint Prediction for Rectified Flow ModelsDiffusion models → Flow matching, time series and science
Token Reduction Is Not Cost ReductionLLM agents and tool use → Agent memory systems +1
An Agentic AI Scientific Community for Automated Neural Operator DiscoveryLLM agents and tool use → Web, search, and research agents
FlashDiff: Efficient Regional Execution and Scheduling for Diffusion Model ServingDiffusion models → Diffusion foundations and sampling
Continual Learning with Elastic Regularization and Synthetic Replay for Federated MLLM Fine-TuningMultimodal LLMs / vision-language models → Prompt learning and adaptation of CLIP
PFAdapter: Hierarchical LoRA Decomposition for Personalized Federated MLLMsLLM agents and tool use → Agentic AI for wireless networks
TraceSynth: Generating Production-Quality Kernel Traces with Constraint-Guided Diffusion ModelsDiffusion models → Flow matching, time series and science
CityBehavEx: A Scalable and Empirically Validated LLM-Assisted Urban Simulation PlatformLLM agents and tool use → Multimodal and embodied agents
Operationalising Multi-Dimensional Evaluation for Conversational Agents: A Scalable, Governed Pipeline with Selective Re-evaluation and Model BenchmarkingLLM agents and tool use → General LLM agent frameworks and benchmarks
The Capacity of Thought: Benchmarking Llama 3.2 in Semantic fMRI Neural Language Decoding and Improving the Huth Encoding-Model BaselineLLM reasoning → Reasoning benchmarks and evaluation
Beyond Test Presence: Assessing the Quality and Robustness of Agent-Generated Tests in Open-Source ProjectsLLM agents and tool use → Software engineering coding agents
AutoTrace: From Patches to Triggers via Agentic Interprocedural ExplorationLLM agents and tool use → Agent threats, guardrails and governance
Designing Agent-Ready Websites for AI Web Agents: A Framework for Machine Readability, Actionability, and Decision ReliabilityLLM agents and tool use → Web, search, and research agents
Agentic systems for breast cancer treatment recommendationsLLM agents and tool use → Medical and scientific domain agents
EFLUX: Elastic Multi-Robot Formation Navigation and Adaptation with Agentic LLMsLLM agents and tool use → Multimodal and embodied agents
SymbOmni: Evolving Agentic Omni Models via Symbolic Concept LearningLLM agents and tool use → Multimodal and embodied agents
Underwriting the Agent Economy: The Blueprint for an AI Insurance StackLLM agents and tool use → Financial trading and forecasting agents
Signal-Guided Optimization for Machine UnlearningLLM safety and security → Machine unlearning
Read It Back: Pretrained MLLMs Are Zero-Shot Reward Models for Text-to-Image GenerationLLM alignment / RLHF / preference optimisation → Diffusion and multimodal model alignment
Requential Coding: Pushing the Limits of Model Compression with Self-Generated Training DataSynthetic data and self-improvement for LLMs → Model collapse from synthetic data
Invariant Learning Dynamics of Transformers in Inductive Reasoning TasksLLM reasoning → Prompting and chain-of-thought reasoning
Beyond the Single Camera: Agentic Multi-View Reasoning in Sports Video UnderstandingLLM reasoning → Multimodal reasoning +1
Cycle-World: Mitigating Error Accumulation in Long-term Video World Models via Reverse-Prediction Cycle ConsistencyWorld models → Video-generation world models and simulators +1
MM-ToolSandBox: A Unified Framework for Evaluating Visual Tool-Calling AgentsLLM agents and tool use → Multimodal and embodied agents
An Exact Instrument for State Usage in Selective State-Space Models, and the Input-Driven Migration It RevealsState-space sequence models (S4, Mamba) → Core SSM architectures for language
How Temperature Shapes Ideological Discourse in Retrieval-Augmented Generation?Retrieval-augmented generation → RAG robustness and adaptive retrieval
From Global to Factor-Wise Expert Composition in Discrete Diffusion ModelsDiffusion models → Diffusion foundations and sampling
When Local Monitors Miss Compositional Harm: Diagnosing Distributed Backdoors in Multi-Agent SystemsLLM agents and tool use → Agent threats, guardrails and governance +1
MET: Theory-Grounded and Culture-Aware Multilingual Moral ReasoningLLM reasoning → Reasoning benchmarks and evaluation
Active Offline-to-Online Reinforcement LearningOffline RL and imitation learning from data → Offline RL algorithms
An Explainable Agentic System for Detection of Conversational Scams with Summary-Based MemoryLLM agents and tool use → Agent threats, guardrails and governance
Agent Hacks Agent: Autoresearch for Production-Agent Red-TeamingLLM agents and tool use → Agent threats, guardrails and governance +1
Think Through a Bottleneck: Hourglass Reasoning for Rigorous InductionLLM reasoning → Reasoning benchmarks and evaluation
From World Action Models to Embodied Brains: A Roadmap for Open-World Physical IntelligenceWorld models → World models for navigation
RAGU: A Multi-Step GraphRAG Engine with a Compact Domain-Adapted LLMRetrieval-augmented generation → Knowledge-graph and structured-knowledge RAG
ABot-3DWorld 0: A Universal World Model to Explore Any 3D SpaceWorld models → Video-generation world models and simulators
Feature-Space Guided Diffusion for Realistic Ultrasound Image SynthesisDiffusion models → Medical imaging and segmentation
Xiaomi-Robotics-U0: Unified Embodied Synthesis with World Foundation ModelWorld models → Embodied and robot world models
Lesioned Multimodal Language Models Reproduce Aphasic Picture-Naming PatternsMultimodal LLMs / vision-language models → Medical vision-language models
ThinkLog: Leveraging Reasoning for Log Statement GenerationLLM reasoning → Reasoning for agents, planning and code
Interaction Scaling: Grounding the Third Axis of Test-Time ComputeLLM reasoning → Test-time compute and efficient thinking
HCRMap: Pressure-Aware Hot-Expert Residency Mapping for 3.5D MoE Chiplet InferenceMixture-of-experts → Efficient MoE training and inference systems
Actor as Its Own Critic: Unifying Region Understanding and Localization via CycleGRPOMultimodal LLMs / vision-language models → Embodied and 3D vision-language models
DiffEEG: A Self-Supervised Denoising Diffusion Model for Learning EEG Generic RepresentationsDiffusion models → Medical imaging and segmentation
Heuristic Learning for Active Flow Control Using Coding AgentsLLM agents and tool use → Reinforcement learning for LLM agents
PaperRouter-Agent: A Content-Grounded LLM Agent for Personalized Hierarchical Paper RoutingLLM agents and tool use → Web, search, and research agents
Adaptive Routing for Efficient Diffusion Transformer-Based PNI PredictionDiffusion models → Medical imaging and segmentation
Parse, Search, and Confirmation: Training-Free Aerial Vision-and-Dialog Navigation with Chain-of-Thought Reasoning and Structured Spatial MemoryLLM reasoning → Spatial and embodied reasoning
SCOPE-RL: Optimizing Reasoning Paths Before and After SuccessLLM reasoning → Reinforcement learning for reasoning
GEIS: A Generation-Evaluation-Improvement Loop of Agent Skills for Long-Form Article GenerationLLM agents and tool use → Web, search, and research agents
See like a Robot: Robot-Centric Pointmaps for VLA ModelsLLM reasoning → Spatial and embodied reasoning +1
Agentic Skill Optimization over Lie AlgebroidsLLM agents and tool use → Agent skill learning and evolution
HyperSafe: Inference-Time Safety Recovery for Fine-Tuned Language ModelsLLM alignment / RLHF / preference optimisation → Safety alignment and jailbreak attacks +1
FAIR GraphRAG: A Retrieval-Augmented Generation Approach for Semantic Data AnalysisRetrieval-augmented generation → Knowledge-graph and structured-knowledge RAG
UMoE:Unlocking Every Expert in Domain-Specific TrainingMixture-of-experts → MoE large language models
Velocity Scheduled Flow MatchingDiffusion models → Flow matching, time series and science
The Ebb and Flow of Multimodal Focus: Scheduling Visual Relay Windows for Grounded VLM ReasoningLLM reasoning → Multimodal reasoning +1
ToFu: A White-Box, Token-Efficient Agent Harness for ResearchersLLM agents and tool use → Software engineering coding agents
Cross-Architecture LLM Ensembles, Feature-Based Reranking and Retrieval-Augmented Prompting for Legal Information ProcessingRetrieval-augmented generation → Legal and financial RAG
Agentic Routing: The Harness-Native Data FlywheelLLM agents and tool use → General LLM agent frameworks and benchmarks
TerraRepair: A Tool-Grounded LLM Agent for Infrastructure-as-Code RepairLLM agents and tool use → Software engineering coding agents
StructAgent: Harness Long-horizon Digital Agents with Unified Causal StructureMultimodal LLMs / vision-language models → Embodied and 3D vision-language models +1
OpsMem: Dual-Memory Reasoning with Cross-Memory Resonance for Failure DiagnosisLLM reasoning → Reasoning for agents, planning and code
Compile, Then Page: Executable SOP Programs and a Capability-Gated Runtime for Procedural LLM AgentsLLM agents and tool use → Software engineering coding agents
Calibrated e-CUSUM Decoding for Quantized Reasoning Models: Why Token Log-Probability Is the Wrong Observable for Decoding MonitorsLLM reasoning → Test-time compute and efficient thinking
FlowArk: Boosting Agentic Data-flow Analysis for Android Apps via Context-Aware Knowledge ReuseLLM agents and tool use → Software engineering coding agents
Efficient Test-Time Optimization for Multi-Agent Proof AutoformalizationLLM reasoning → Mathematical reasoning
Mako: A Self-Evolving Agentic Operating System (SE-AOS) for Autonomous Web ExploitationLLM agents and tool use → Agent threats, guardrails and governance
Automated Textbook Auditing with Multi-Agent LLM SystemsLLM agents and tool use → Web, search, and research agents
Towards Predictive, Aligned, and Scalable Robot LearningWorld models → Embodied and robot world models
Enhancing LLMs through human feedback: a journey towards self-improvementRetrieval-augmented generation → RAG pipeline design and evaluation +1
Valid ne Necessary: Diagnosing Latent Inefficiency in Chain-of-ThoughtLLM reasoning → Prompting and chain-of-thought reasoning
TreeThink: A Modular Tree Search Library for Mathematical Reasoning with LLMsLLM reasoning → Mathematical reasoning
Multi-Agent LLMs Fail to Explore Each OtherLLM agents and tool use → General LLM agent frameworks and benchmarks
DeepBias: Adaptive In-depth Probing of Social Biases in LVLMsMultimodal LLMs / vision-language models → Social bias and fairness in VLMs
Heterogeneous Agent Cohorts for Safe Open-Ended Exploration with Runtime Constraint MemoryLLM agents and tool use → Agent threats, guardrails and governance
HandFlow: Fully Generative 4D Hand Recovery with Flow MatchingDiffusion models → Diffusion and flow policies for robotics
ProgramTab: Boosting Table Reasoning of LLMs via Programmatic ParadigmLLM reasoning → Knowledge- and tool-augmented reasoning
Slot-RAE: Streamlining Object-Centric Learning via Direct Representation Auto-EncodersDiffusion models → Image generation, text-to-image and editing
SCALECUA: Scaling Computer Use Agents with Verifiable Task Synthesis and Efficient Online RLLLM agents and tool use → GUI and mobile agents
The Path to Self-Evolving Clinical Systems: Scaling Medical Agents from Assistance to AutonomyMultimodal LLMs / vision-language models → Medical vision-language models +2

All 429