Papers

New on arXiv

Sat 2 May 2026

234 papers

BaRA: Budget-constrained and Reliable Web Data Collection AgentLLM agents and tool use → Web, search, and research agents
Ghost in the Context: Policy-Carriage Integrity in LLM AgentsLLM agents and tool use → Agent threats, guardrails and governance
NoiseRater: Meta-Learned Noise Valuation for Diffusion Model TrainingDiffusion models → Diffusion foundations and sampling
Reasoning emerges from constrained inference manifolds in large language modelsLLM reasoning → Reasoning benchmarks and evaluation
DataArc-SynData-Toolkit: A Unified Closed-Loop Framework for Multi-Path, Multimodal, and Multilingual Data SynthesisSynthetic data and self-improvement for LLMs → LLM synthetic data generation
Tracing the Dynamics of Refusal: Exploiting Latent Refusal Trajectories for Robust Jailbreak DetectionLLM safety and security → Jailbreaking
Toward a Principled Framework for Agent Safety MeasurementLLM agents and tool use → Agent threats, guardrails and governance
Adaptive Pluralistic Alignment: A pipeline for dynamic artificial democracyLLM alignment / RLHF / preference optimisation → Human values and AI feedback alignment
Less Interaction But More Explanation: A Communication Perspective on Agentic AI InterfacesLLM agents and tool use → General LLM agent frameworks and benchmarks
Concepts Whisper: Spectral Anti-Concentration and the Dual Geometry of Transformer RepresentationsMechanistic interpretability → Sparse autoencoder training and evaluation
Evaluating Agentic AI in the Wild: Failure Modes, Drift Patterns, and a Production Evaluation FrameworkLLM agents and tool use → General LLM agent frameworks and benchmarks
Hyper-DP3: Frequency-Aware Right-Sizing of 3D Diffusion Policies for Visuomotor ControlDiffusion models → Diffusion and flow policies for robotics
Hybrid Quantum Reinforcement Learning with QAOA for Improved Vehicle Routing OptimizationDeep reinforcement learning methods → Quantum reinforcement learning
Unifying Deep Stochastic Processes for Image EnhancementDiffusion models → Diffusion foundations and sampling
Feedback-Normalized Developer Memory for Reinforcement-Learning Coding Agents: A Safety-Gated MCP ArchitectureLLM agents and tool use → Agent memory systems
Multi-Agent Reasoning Improves Compute Efficiency: Pareto-Optimal Test-Time ScalingLLM reasoning → Reasoning for agents, planning and code +1
Multi-Dataset Cross-Domain Knowledge Distillation for Unified Medical Image Segmentation, Classification, and DetectionKnowledge distillation → Distillation for medical imaging
Neuro-Symbolic Agents for Hallucination-Free Requirements ReuseLLM agents and tool use → Software engineering coding agents
Automated Interpretability and Feature Discovery in Language Models with AgentsLLM agents and tool use → General LLM agent frameworks and benchmarks +1
6G Needs Agents: Toward Agentic AI-Native Networks for Autonomous IntelligenceLLM agents and tool use → Agentic AI for wireless networks
An Efficient Metric for Data Quality Measurement in Imitation LearningOffline RL and imitation learning from data → Robot manipulation imitation learning
Good in Bad (GiB): Sifting Through End-user Demonstrations for Learning a Better PolicyOffline RL and imitation learning from data → Robot manipulation imitation learning
MIRL: Mutual Information-Guided Reinforcement Learning for Vision-Language ModelsMultimodal LLMs / vision-language models → Embodied and 3D vision-language models
Dynamics Distillation for Efficient and Transferable Control LearningDeep reinforcement learning methods → RL for autonomous driving and traffic
MelShield: Robust Mel-Domain Audio Watermarking for Provenance Attribution of AI Generated Synthesized SpeechLLM safety and security → Watermarking generated content and models
Protein-Conditioned Multi-Objective Reinforcement Learning for Full-Length mRNA DesignDeep reinforcement learning methods → RL for energy, networks and operations
SwiftPie: Lightning-fast Subject-driven Image Personalization via One step DiffusionDiffusion models → Image generation, text-to-image and editing
MILD: Mediator Agent System with Bidirectional Perception and Multi-Layered Alignment for Human-Vehicle CollaborationLLM alignment / RLHF / preference optimisation → Reward modeling and RLHF
FT-RAG: A Fine-grained Retrieval-Augmented Generation Framework for Complex Table ReasoningLLM reasoning → Knowledge- and tool-augmented reasoning +1
SciResearcher: Scaling Deep Research Agents for Frontier Scientific ReasoningLLM reasoning → Knowledge- and tool-augmented reasoning +1
Evaluating LLMs on Large-Scale Graph Property Estimation via Random WalksLLM reasoning → Knowledge- and tool-augmented reasoning
Grounding Multi-Hop Reasoning in Structural Causal Models via Group Relative Policy OptimizationLLM reasoning → Reasoning benchmarks and evaluation
AttnRouter: Per-Category Attention Routing for Training-Free Image Editing on MMDiTDiffusion models → Image generation, text-to-image and editing
CSGuard: Toward Forgery-Resistant Watermarking in Diffusion Models via Compressed Sensing ConstraintDiffusion models → Robustness, privacy and adversarial aspects +1
LIE: LiDAR-only HD Map Construction with Intensity Enhancement via Online Knowledge DistillationKnowledge distillation → Detection and dense prediction distillation
Action Agent: Agentic Video Generation Meets Flow-Constrained DiffusionDiffusion models → Diffusion and flow policies for robotics +1
ReMedi: Reasoner for Medical Clinical PredictionLLM reasoning → Clinical and biomedical applications
Practical Limits of Autonomous Test Repair: A Multi-Agent Case Study with LLM-Driven Discovery and Self-CorrectionLLM agents and tool use → Software engineering coding agents
LocalAlign: Enabling Generalizable Prompt Injection Defense via Generation of Near-Target Adversarial Examples for Alignment TrainingLLM safety and security → Prompt injection and agent security
CoFlow: Coordinated Few-Step Flow for Offline Multi-Agent Decision MakingOffline RL and imitation learning from data → Offline multi-agent RL
VisInject: Disruption != Injection -- A Dual-Dimension Evaluation of Universal Adversarial Attacks on Vision-Language ModelsMultimodal LLMs / vision-language models → MLLM jailbreaks and adversarial robustness
Who Decides What Is Harmful? Content Moderation Policy Through A Multi-Agent Personalised Inference FrameworkLLM agents and tool use → Agent threats, guardrails and governance
Injecting Distributional Awareness into MLLMs via Reinforcement Learning for Deep Imbalanced RegressionMultimodal LLMs / vision-language models → Embodied and 3D vision-language models
Verbal-R3: Verbal Reranker as the Missing Bridge between Retrieval and ReasoningLLM reasoning → Knowledge- and tool-augmented reasoning +1
LiveFMBench: Unveiling the Power and Limits of Agentic Workflows in Specification GenerationLLM agents and tool use → Software engineering coding agents
VISTA: Video Interaction Spatio-Temporal Analysis BenchmarkMultimodal LLMs / vision-language models → Embodied and 3D vision-language models
MemORAI: Memory Organization and Retrieval via Adaptive Graph Intelligence for LLM Conversational AgentsLLM agents and tool use → Agent memory systems
MTA: Multi-Granular Trajectory Alignment for Large Language Model DistillationLLM alignment / RLHF / preference optimisation → LLM preference optimization methods +1
ESARBench: A Benchmark for Agentic UAV Embodied Search and RescueLLM agents and tool use → Multimodal and embodied agents
PACE: Parameter Change for Unsupervised Environment DesignDeep reinforcement learning methods → Deep RL algorithms and representations
AgriKD: Cross-Architecture Knowledge Distillation for Efficient Leaf Disease ClassificationKnowledge distillation → Detection and dense prediction distillation
LLM Output Detectability and Task Performance Can be Jointly OptimizedLLM safety and security → Watermarking generated content and models
MAD-OPD: Breaking the Ceiling in On-Policy Distillation via Multi-Agent DebateLLM agents and tool use → Reinforcement learning for LLM agents +1
The Perceptual Bandwidth Bottleneck in Vision-Language Models: Active Visual Reasoning via Sequential Experimental DesignLLM reasoning → Multimodal reasoning +1
OralMLLM-Bench: Evaluating Cognitive Capabilities of Multimodal Large Language Models in Dental PracticeMultimodal LLMs / vision-language models → Medical vision-language models
Truth or Tribe: How In-group Favoritism Prioritize Facts in Persona AgentsLLM agents and tool use → Social simulation and multi-agent debate
Segment-Aligned Policy Optimization for Multi-Modal ReasoningLLM reasoning → Reinforcement learning for reasoning
Rethinking Model Selection in VLM Through the Lens of Gromov-Wasserstein DistanceLLM reasoning → Multimodal reasoning +1
Beyond Perceptual Shortcuts: Causal-Inspired Debiasing Optimization for Generalizable Video Reasoning in Lightweight MLLMsLLM reasoning → Multimodal reasoning +1
Beyond Semantic Relevance: Counterfactual Risk Minimization for Robust Retrieval-Augmented GenerationRetrieval-augmented generation → RAG robustness and adaptive retrieval
Lifting Traces to Logic: Programmatic Skill Induction with Neuro-Symbolic Learning for Long-Horizon Agentic TasksLLM agents and tool use → Agent skill learning and evolution
Bi-Level Reinforcement Learning Control for an Underactuated Blimp via Center-of-Mass ReconfigurationDeep reinforcement learning methods → RL for robot navigation and UAV control
Chain of Evidence: Pixel-Level Visual Attribution for Iterative Retrieval-Augmented GenerationRetrieval-augmented generation → Multimodal RAG
Position: LLM Serving Needs Mathematical Optimization and Algorithmic Foundations, Not Just HeuristicsEfficient LLM inference and serving → Serving systems and request scheduling
Valley3: Scaling Omni Foundation Models for E-commerceLLM reasoning → Multimodal reasoning +1
Exploring Prompt Alignment with Clinical Factors in Zero-Shot Segmentation VLMs for NSCLC Tumor SegmentationMultimodal LLMs / vision-language models → Medical vision-language models
FeedbackLLM: Metadata driven Multi-Agentic Language Agnostic Test Case Generator with Evolving prompt and Coverage FeedbackLLM agents and tool use → Software engineering coding agents
S3-R1: Learning to Retrieve and Answer Step-by-Step with Synthetic DataLLM reasoning → Knowledge- and tool-augmented reasoning +1
Breaking the Computational Barrier: Provably Efficient Actor-Critic for Low-Rank MDPsDeep reinforcement learning methods → RL theory and sample complexity
Rhamba: Region-Aware Hybrid Attention-Mamba Framework for Self-Supervised Learning in Resting-State fMRIState-space sequence models (S4, Mamba) → Physiological signal modeling
A Principled Approach for Creating High-fidelity Synthetic Demonstrations for Imitation LearningOffline RL and imitation learning from data → Robot manipulation imitation learning
Local Hessian Spectral Filtering for Robust Intrinsic Dimension EstimationDiffusion models → Diffusion foundations and sampling
Agentic AI Systems Should Be Designed as Marginal Token AllocatorsLLM agents and tool use → Agent memory systems
ClarifySTL: An Interactive LLM Agent Framework for STL Transformation through Requirements ClarificationLLM agents and tool use → Software engineering coding agents
Faithful Mobile GUI Agents with Guided Advantage EstimatorLLM agents and tool use → GUI and mobile agents
SRA: Span Representation Alignment for Large Language Model DistillationLLM alignment / RLHF / preference optimisation → LLM preference optimization methods +1
GR-Ben: A General Reasoning Benchmark for Evaluating Process Reward ModelsLLM reasoning → Reinforcement learning for reasoning +1
To Do or Not to Do: Ensuring the Safety of Visuomotor Policies Learned from DemonstrationsOffline RL and imitation learning from data → Safe offline RL
TAIL-Safe: Task-Agnostic Safety Monitoring for Imitation Learning PoliciesDiffusion models → Diffusion and flow policies for robotics +1
VLA-ATTC: Adaptive Test-Time Compute for VLA Models with Relative Action Critic ModelLLM reasoning → Spatial and embodied reasoning +1
Sentinel-VLA: A Metacognitive VLA Model with Active Status Monitoring for Dynamic Reasoning and Error RecoveryVision-language-action models → Efficient VLA inference and adaptation
Phase-map synthesis from magnitude-only MR images using conditional score-based diffusion models with application in training of accelerated MRI reconstruction modelsDiffusion models → Medical imaging and segmentation
Hybrid LSTM-Graph Neural Framework for Robust Financial Fraud Detection and Adversarial Resilience
Quantum Compositional NLP for Arabic: Grammar, Morphology, and Word Sense in Circuit Topology
Format Sensitivity Index: Token-Controlled Prompt Wrapper Robustness and Schema Compliance in LLM Benchmarking
From Awareness to Action: Understanding and Overcoming the Research-Practice Gap in Algorithmic Fairness for Public Health
AI-Integrated Learning Management System for Middle School: A Longitudinal Study of Learning Outcomes Through High School and Beyond
Right-Sizing Communication and Recommendation Set Size in AI-Assisted Search
Quantization Undoes Alignment: Bias Emergence in Compressed LLMs Across Models and Precision Levels
BioSEN: A Bio-acoustic Signal Enhancement Network for Animal Vocalizations
HoReN: Normalized Hopfield Retrieval for Large-Scale Sequential Model Editing
Forecasting Source Stability in Scientific Experiments using Temporal Learning Models: A Case Study from Tritium Monitoring
Intelligent Autonomous Orchestration for Distributed Cloud Resources using Complex-Stability Analysis
Weight Pruning Amplifies Bias: A Multi-Method Study of Compressed LLMs for Edge AI
State Representation and Termination for Recursive Reasoning Systems
Dynamic Graph with Similarity-Aware Attention Graph Neural Network for Recommender Systems
Regularized Centered Emphatic Temporal Difference Learning
EFGPP: Exploratory framework for genotype-phenotype prediction
AI Alignment via Incentives and Correction
Prescriptive Scaling Laws for Data Constrained Training

All 234