Papers

New on arXiv

Sat 18 Apr 2026

239 papers

FlowEvo: Self-Evolving Agents through the Co-Evolution of Workflows and Executable SkillsLLM agents and tool use → General LLM agent frameworks and benchmarks +1
StoryMI: Steerable Multi-Agent Therapeutic Dialogue GenerationLLM agents and tool use → Medical and scientific domain agents
LFRAG: Layout-oriented Fine-grained Retrieval-Augmented Generation on Multimodal Document UnderstandingRetrieval-augmented generation → Multimodal RAG
HealthCraft: A Reinforcement Learning Safety Environment for Emergency MedicineDeep reinforcement learning methods → Safe and constrained RL
Do LLMs Hold Their Values? MANTA: A Multi-Turn Adversarial Benchmark for Animal Welfare ReasoningLLM reasoning → Reasoning benchmarks and evaluation
GraphDC: A Divide-and-Conquer Multi-Agent System for Scalable Graph Algorithm ReasoningLLM reasoning → Knowledge- and tool-augmented reasoning +1
Structural Instability of Feature CompositionMechanistic interpretability → Sparse autoencoder training and evaluation
PivotMerge: Bridging Heterogeneous Multimodal Pre-training via Post-Alignment Model MergingMultimodal LLMs / vision-language models → MLLM training and visual instruction tuning
DO-Bench: An Attributable Benchmark for Diagnosing Object Hallucination in Vision-Language ModelsMultimodal LLMs / vision-language models → Hallucination mitigation in LVLMs
Beyond Task Success: An Evidence-Synthesis Framework for Evaluating, Governing, and Orchestrating Agentic AILLM agents and tool use → Agent threats, guardrails and governance
Can We Build Scene Graphs, Not Classify Them? FlowSG: Progressive Image-Conditioned Scene Graph Generation with Flow MatchingDiffusion models → Image generation, text-to-image and editing
Systematic Capability Benchmarking of Frontier Large Language Models for Offensive Cyber TasksLLM agents and tool use → Agent threats, guardrails and governance
FlowRefiner: Flow Matching-Based Iterative Refinement for 3D Turbulent Flow SimulationDiffusion models → Flow matching, time series and science
Graph-of-Agents: A Graph-based Framework for Multi-Agent LLM CollaborationLLM agents and tool use → General LLM agent frameworks and benchmarks
Logic-Based Verification of Task Allocation for LLM-Enabled Multi-Agent Manufacturing SystemsLLM agents and tool use → General LLM agent frameworks and benchmarks
The Consensus Trap: Rescuing Multi-Agent LLMs from Adversarial Majorities via Token-Level CollaborationLLM agents and tool use → Social simulation and multi-agent debate
Please refuse to answer me! Mitigating Over-Refusal in Large Language Models via Adaptive Contrastive DecodingLLM alignment / RLHF / preference optimisation → Safety alignment and jailbreak attacks +1
Prompt Sensitivity in Vision-Language Grounding: How Small Changes in Wording Affect Object DetectionMultimodal LLMs / vision-language models → Prompt learning and adaptation of CLIP
CASCADE: A Component Ablation and Corpus Audit of a Layered Local Defense for MCP-Based SystemsLLM agents and tool use → Agent threats, guardrails and governance
The Provenance Gap in Clinical AI: Evidence-Traceable Temporal Knowledge Graphs for Rare Disease ReasoningLLM reasoning → Clinical and biomedical applications
Complementing Self-Consistency with Cross-Model Disagreement for Uncertainty QuantificationLLM reasoning → Test-time compute and efficient thinking
HiveMind: OS-Inspired Scheduling for Concurrent LLM Agent WorkloadsLLM agents and tool use → Software engineering coding agents
Live LTL Progress Tracking: Towards Task-Based ExplorationDeep reinforcement learning methods → Interpretable and logic-guided RL
HarmChip: Evaluating Hardware Security Centric LLM Safety via Jailbreak BenchmarkingLLM safety and security → Safety alignment, refusal and guardrails
GenericAgent: A Token-Efficient Self-Evolving LLM Agent via Contextual Information Density Maximization (V1.0)LLM agents and tool use → Agent memory systems +1
EvoComp: Learning Visual Token Compression for Multimodal Large Language Models via Semantic-Guided Evolutionary LabelingMultimodal LLMs / vision-language models → Visual token pruning and efficiency +1
Abstain-R1: Calibrated Abstention and Post-Refusal Clarification via Verifiable RLLLM reasoning → Test-time compute and efficient thinking
Stability-Weighted Decoding for Diffusion Language ModelsDiffusion models → Discrete and text diffusion
Jailbreaking Large Language Models with Morality AttacksLLM safety and security → Safety alignment, refusal and guardrails
OASIS: On-Demand Hierarchical Event Memory for Streaming Video ReasoningLLM reasoning → Multimodal reasoning
Web-Gewu: A Browser-Based Interactive Playground for Robot Reinforcement LearningDeep reinforcement learning methods → RL for robot locomotion and manipulation
SIF: Semantically In-Distribution Fingerprints for Large Vision-Language ModelsMultimodal LLMs / vision-language models → MLLM jailbreaks and adversarial robustness
Harness as an Asset: Enforcing Determinism via the Convergent AI Agent Framework (CAAF)LLM agents and tool use → Software engineering coding agents
Beyond Static Benchmarks: Synthesizing Harmful Content via Persona-based Simulation for Robust EvaluationLLM safety and security → Safety alignment, refusal and guardrails
Improving LLM Code Reasoning via Semantic Equivalence Self-Play with Formal VerificationLLM reasoning → Reasoning benchmarks and evaluation
Bolzano: Case Studies in LLM-Assisted Mathematical ResearchLLM agents and tool use → Educational agents and math problem solving
DVAR: Adversarial Multi-Agent Debate for Video Authenticity DetectionLLM agents and tool use → Multimodal and embodied agents
DOSE: Data Selection for Multi-Modal LLMs via Off-the-Shelf ModelsMultimodal LLMs / vision-language models → MLLM training and visual instruction tuning
MCPO: Mastery-Consolidated Policy Optimization for Large Reasoning ModelsLLM reasoning → Reinforcement learning for reasoning
On Safety Risks in Experience-Driven Self-Evolving AgentsLLM agents and tool use → Agent threats, guardrails and governance +1
NaviFormer: A Deep Reinforcement Learning Transformer-like Model to Holistically Solve the Navigation ProblemDeep reinforcement learning methods → RL for robot navigation and UAV control
Visual Inception: Compromising Long-term Planning in Agentic Recommenders via Multimodal Memory PoisoningLLM agents and tool use → Agent threats, guardrails and governance
Self-Reasoning Agentic Framework for Narrative Product Grid-Collage GenerationLLM agents and tool use → Multimodal and embodied agents
Open-TQ-Metal: Fused Compressed-Domain Attention for Long-Context LLM Inference on Apple SiliconEfficient LLM inference and serving → KV cache optimization
TSM-Pose: Topology-Aware Learning with Semantic Mamba for Category-Level Object Pose EstimationState-space sequence models (S4, Mamba) → Video, motion and generative Mamba models
MNAFT: modality neuron-aware fine-tuning of multimodal large language models for image translationMultimodal LLMs / vision-language models → MLLM training and visual instruction tuning
MEMRES: A Memory-Augmented Resolver with Confidence Cascade for Agentic Python Dependency ResolutionLLM agents and tool use → Agent memory systems
Playing Psychic: Using Thought Trees to Predict Reasoning Models Accuracy on Coding TasksLLM reasoning → Test-time compute and efficient thinking
CoGR-MoE: Concept-Guided Expert Routing with Consistent Selection and Flexible Reasoning for Visual Question AnsweringMixture-of-experts → Multimodal MoE models
MeasHalu: Mitigation of Scientific Measurement Hallucinations for Large Language Models with Enhanced ReasoningLLM reasoning → Hallucination detection and mitigation
Alignment Imprint: Zero-Shot AI-Generated Text Detection via Provable Preference DiscrepancyLLM alignment / RLHF / preference optimisation → LLM preference optimization methods
ClimAgent: LLM as Agents for Autonomous Open-ended Climate Science AnalysisLLM agents and tool use → Multimodal and embodied agents
Noise-Adaptive Diffusion Sampling for Inverse Problems Without Task-Specific TuningDiffusion models → Inverse problems and image restoration
Freshness-Aware Prioritized Experience Replay for LLM/VLM Reinforcement LearningLLM reasoning → Reinforcement learning for reasoning +1
x1: Learning to Think Adaptively Across Languages and CulturesLLM reasoning → Reasoning benchmarks and evaluation
When Choices Become Risks: Safety Failures of Large Language Models under Multiple-Choice ConstraintsLLM alignment / RLHF / preference optimisation → Safety alignment and jailbreak attacks
KIRA: Knowledge-Intensive Image Retrieval and Reasoning Architecture for Specialized Visual DomainsLLM reasoning → Multimodal reasoning +1
The Cognitive Penalty: Ablating System 1 and System 2 Reasoning in Edge-Native SLMs for Decentralized ConsensusLLM reasoning → Reasoning for agents, planning and code
Skilldex: A Package Manager and Registry for Agent Skill Packages with Hierarchical Scope-Based DistributionLLM agents and tool use → Agent skill learning and evolution
PRISM: Probing Reasoning, Instruction, and Source Memory in LLM HallucinationsLLM reasoning → Hallucination detection and mitigation
EasyVideoR1: Easier RL for Video UnderstandingLLM reasoning → Multimodal reasoning
CrossFlowDG: Bridging the Modality Gap with Cross-modal Flow Matching for Domain GeneralizationDiffusion models → Flow matching, time series and science
Step-GRPO: Internalizing Dynamic Early Exit for Efficient ReasoningLLM reasoning → Test-time compute and efficient thinking +1
Prune, Interpret, Evaluate: A Cross-Layer Transcoder-Native Framework for Efficient Circuit Discovery via Feature AttributionMechanistic interpretability → Circuits and mechanistic analysis
SinkRouter: Sink-Aware Routing for Efficient Long-Context Decoding in Large Language and Multimodal ModelsEfficient LLM inference and serving → KV cache optimization
GRAIL: Autonomous Concept Grounding for Neuro-Symbolic Reinforcement LearningLLM reasoning → Reinforcement learning for reasoning +1
Governed MCP: Kernel-Level Tool Governance for AI Agents via Logit-Based Safety PrimitivesLLM agents and tool use → Agent threats, guardrails and governance
HieraSparse: Hierarchical Semi-Structured Sparse KV AttentionEfficient LLM inference and serving → KV cache optimization
Learning to Trade Like an Expert: Cognitive Fine-Tuning for Stable Financial Reasoning in Language ModelsLLM reasoning → Reasoning for agents, planning and code
GAMMA-Net: Adaptive Long-Horizon Traffic Spatio-Temporal Forecasting Model based on Interleaved Graph Attention and Multi-Axis MambaState-space sequence models (S4, Mamba) → Video, motion and generative Mamba models
Q-DeepSight: Incentivizing Thinking with Images for Image Quality Assessment and RefinementLLM reasoning → Multimodal reasoning
Refinement of Accelerated Demonstrations via Incremental Iterative Reference Learning Control for Fast Contact-Rich Imitation LearningOffline RL and imitation learning from data → Robot manipulation imitation learning
HeLa-Mem: Hebbian Learning and Associative Memory for LLM AgentsLLM agents and tool use → Agent memory systems
The Illusion of Certainty: Decoupling Capability and Calibration in On-Policy DistillationKnowledge distillation → On-policy and self-distillation for LLMs
SafeDream: Safety World Model for Proactive Early Jailbreak DetectionWorld models → World model security and GUI agents +2
Online Reinforcement Learning for Safe Gain Scheduling in Nonlinear Quadrotor ControlDeep reinforcement learning methods → RL for robot navigation and UAV control
PersonalHomeBench: Evaluating Agents in Personalized Smart HomesLLM agents and tool use → Financial trading and forecasting agents
Channel Attention-Guided Cross-Modal Knowledge Distillation for Referring Image SegmentationKnowledge distillation → Detection and dense prediction distillation
Bridging Coarse and Fine Recognition: A Hybrid Approach for Open-Ended Multi-Granularity Object Recognition in Interactive Educational GamesMultimodal LLMs / vision-language models → VLM benchmarks and visual reasoning
Federation over Text: Insight Sharing for Multi-Agent ReasoningLLM reasoning → Reasoning for agents, planning and code +1
Privacy-Aware Machine Unlearning with SISA for Reinforcement Learning-Based Ransomware DetectionLLM safety and security → Machine unlearning
Can AI Evaluate AI Scientists? A Benchmarking Study of Autonomous Research Generation Systems Using Automated Multi-Model Review
From "Strings" to "Things" for Personal Knowledge Graphs: Evaluating LLM Triple Extraction for Recommendation Systems
LANTERN: Layered Archival and Temporal Episodic Retrieval Network for Long-Context LLM Conversations
Multi-Granularity Reasoning for Natural Language Inference
From Scoring to Explanations: Evaluating SHAP and LLM Rationales for Rubric-based Teaching Quality Assessment
Efficient Punctuation Restoration via Weighted Lookahead Scoring Method for Streaming ASR Systems
One Mask to Rule Them All: On Hidden Facts after Editing and How to Find Them
Spectral Analysis of Fake News Propagation
Channel-Level Semantic Perturbations: Unlearnable Examples for Diverse Training Paradigms
Adaptive Computation Depth via Learned Token Routing in Transformers
TimeRFT: Stimulating Generalizable Time Series Forecasting for TSFMs via Reinforcement Finetuning
Emergence Transformer: Dynamical Temporal Attention Matters
StomaD2: An All-in-One System for Intelligent Stomatal Phenotype Analysis via Diffusion-Based Restoration Detection Network
Vision-Based Human Awareness Estimation for Enhanced Safety and Efficiency of AMRs in Industrial Warehouses
Quantum AI for Cancer Diagnostic Biomarker Discovery
The Virtue of Sparsity in Complexity
PPEDCRF: Dynamic-CRF-Guided Selective Perturbation for Background-Based Location Privacy in Video Sequences
Lightweight Cybersickness Detection based on User-Specific Eye and Head Tracking Data in Virtual Reality
Uncertainty Quantification in PINNs for Turbulent Flows: Bayesian Inference and Repulsive Ensembles

All 239