Papers

New on arXiv

Sat 4 Jul 2026

188 papers

VFA: Empowering Multilingual MLLMs via Vision-Free AdaptationMultimodal LLMs / vision-language models → MLLM training and visual instruction tuning
Agentao: A Policy-Governed Runtime Harness for Embeddable Tool-Using LLM AgentsLLM agents and tool use → Agent threats, guardrails and governance
Humans are Missing from AI Coding Agent ResearchLLM agents and tool use → Software engineering coding agents
From Survey Personas to LLM Agents: A Generative Agent-based Simulation of Mobility Policy Preference DynamicsLLM agents and tool use → Social simulation and multi-agent debate
Telco-GAIA: Bilingual Benchmark for Agents in Telecom DomainLLM agents and tool use → General LLM agent frameworks and benchmarks
HyGRL: Adaptive Hybrid Graph Reasoning for Multi-Entity QuestionsLLM reasoning → Knowledge- and tool-augmented reasoning +1
HyperDCM: Dynamic Cluster Memory Replay in Hyperbolic Space for Continual Robotic Navigation Across ScenesDiffusion models → Diffusion and flow policies for robotics
Towards Self-Evolving Agents: A Human-Inspired Adaptive Exploration-Exploitation Framework for Genetic Network ProgrammingLLM agents and tool use → General LLM agent frameworks and benchmarks
Unified Backbone Refinement for Diffusion Models via Internal-Latent AnalysisDiffusion models → Diffusion foundations and sampling
Task-Conditioned Synthetic Data Generation for Improving Machine Learning Performance in Agricultural Prediction TasksSynthetic data and self-improvement for LLMs → LLM synthetic data generation
Context Graphs for Proactive Enterprise AgentsRetrieval-augmented generation → Agentic RAG, agents and memory +1
Reward-Gated On-Policy DistillationLLM reasoning → Prompting and chain-of-thought reasoning +1
The "I Don't Know" Filter: Enhancing Agentic Reliability in Function CallingLLM agents and tool use → General LLM agent frameworks and benchmarks
Candidate-Constrained Retrieval-Augmented Generation for LongEval-RAG: System Design and Empirical AnalysisRetrieval-augmented generation → RAG robustness and adaptive retrieval
Refused in Chat, Written in Code: Workflow-Level Jailbreak Construction in IDE Coding AgentsLLM agents and tool use → Agent threats, guardrails and governance +1
Worldscape-MoE: A Unified Mixture-of-Experts World Model for Scalable Heterogeneous Action ControlWorld models → Embodied and robot world models +1
Neuro-Symbolic Reasoning for Vulnerability DetectionLLM reasoning → Reasoning benchmarks and evaluation
Reward Lightning: Fast Video Generation via Homologous Preference DistillationDiffusion models → Video generation +1
NormWorlds-CF: Solver-Verified Counterfactual Normative Reasoning with Metamorphic-Relation GRPOLLM reasoning → Mathematical reasoning
The Remarkable Effectiveness of Providing AI Agents with Natural Language Tools: A Replication Study Validating NLT Performance Across 14 ModelsLLM agents and tool use → General LLM agent frameworks and benchmarks
Online Linear Programming for Multi-Objective Routing in LLM ServingEfficient LLM inference and serving → Serving systems and request scheduling
ThermoForce: A Physics-Structured Interventional World Model for Building HVAC ControlWorld models → Latent predictive world models
MPSelectTune: Prompt-type Selection for Fine-tuning improves Concept Unlearning in LLMsLLM safety and security → Machine unlearning
CDCP: Conditional Diffusion Model with Contextual Prompts for Multi-task Offline Safe Reinforcement LearningDiffusion models → Diffusion and flow policies for robotics +1
USE: A Unified Self-Ensembling Framework for Test-Time Prompt TuningMultimodal LLMs / vision-language models → Prompt learning and adaptation of CLIP
DICT: Data Injection and Contrastive Trajectory Refinement for Conditional Image Generation with Diffusion ModelsDiffusion models → Diffusion foundations and sampling
AdaptiveSD A Stability-Aware, Runtime-Adaptive Speculative Decoding Framework with Multi-Policy Orchestration for CPU-Constrained LLM InferenceEfficient LLM inference and serving → Speculative decoding
High-Fidelity One-Step Generative Visuomotor Policy via Recursive Correction, Frequency Consistency, and Contrastive Flow MatchingDiffusion models → Diffusion and flow policies for robotics
Rethinking Scientific Discovery in the Agentic EraLLM agents and tool use → Web, search, and research agents
CogRad: A Cognitively-Inspired Multi-Agent Framework for Radiology Report GenerationLLM agents and tool use → Medical and scientific domain agents
ContiStain: Cross-Domain Relation-Preserving Distillation for Continual Multi-Domain Virtual IHC StainingMixture-of-experts → Medical and biomedical MoE
When Simpler Is Better: Evaluating Translation Pipelines for Medieval Latin ManuscriptsMultimodal LLMs / vision-language models → Document understanding and multimodal RAG
How Do Diffusion Classifiers Decide? A Bias-Centric EvaluationMultimodal LLMs / vision-language models → Image editing and quality assessment +1
DualView: Preventing Indirect Prompt Injection in Personal AI AgentsLLM agents and tool use → Agent threats, guardrails and governance +1
Global Logic and Local Search: Dual-Stream Multimodal In-Context Learning for Verifiable Industrial Anomaly DetectionMultimodal LLMs / vision-language models → Chart understanding and anomaly detection
CineMobile: On-Device Image-to-Video Diffusion for Cinematic Camera Motion GenerationDiffusion models → Video generation
Tensor-Train Joint Modeling for Few-Step Discrete DiffusionDiffusion models → Diffusion foundations and sampling
Self-Improving Diffusion Classifiers with Minority Preference OptimizationLLM alignment / RLHF / preference optimisation → Diffusion and multimodal model alignment +1
Semantic-aware and Self-improving Program Reduction via Agentic Large Language ModelsLLM agents and tool use → Software engineering coding agents +1
Look Before You Leap: Distilling Tree Search into Action Evaluation for Frozen VLA ModelsVision-language-action models → VLA architectures and generalization
Bridging Interleaved Multi-Modal Reasoning as a Unified Decision ProcessLLM reasoning → Multimodal reasoning
Attending to Multimodal Generation One Token at a TimeMultimodal LLMs / vision-language models → MLLM training and visual instruction tuning
CoGen3D: An Agentic Human-AI Co-Design Pipeline for 3D Asset Generation for Virtual RealityLLM agents and tool use → Multimodal and embodied agents
ProACT: Towards Breakdown-Aware Proactive Agent in Multi-User CollaborationLLM agents and tool use → General LLM agent frameworks and benchmarks
GRASP: Graph-Reasoning Aided Survey Planning for High-Fidelity Related Work GenerationLLM reasoning → Knowledge- and tool-augmented reasoning
Explainable Reinforcement Learning for Adaptive Traffic Signal ControlDeep reinforcement learning methods → RL for autonomous driving and traffic
Agent Reinforcement Learning via Pivotal-Aware Self-Feedback RetryLLM agents and tool use → Reinforcement learning for LLM agents +1
Social Networks of LLM AgentsLLM agents and tool use → Social simulation and multi-agent debate
CoRE-VLA: Towards Scalable and Robust Vision-Language-Action Modeling via Conditional Routing of ExpertsVision-language-action models → Generalist VLA models and benchmarks
Don't Blame the Large Language Model: How Agent Harness Evolution Shapes Coding Agent QualityLLM agents and tool use → Software engineering coding agents
PathMark: Protecting Intellectual Property of Mixture-of-Expert LLMs via Path WatermarksLLM safety and security → Watermarking generated content and models
Diffusion learning reveals viable parameter manifolds and compensation geometry in biological dynamical systemsDiffusion models → Diffusion foundations and sampling
From Geometric Labels to Semantic Understanding of Indoor Building Components Using Multimodal Large Language ModelsMultimodal LLMs / vision-language models → Embodied and 3D vision-language models
ViPo-MLLM: Visual-Pose Multimodal LLM for Gloss-Free Sign Language TranslationMultimodal LLMs / vision-language models → VLM benchmarks and visual reasoning
AutoCedar: An Agentic Framework for Verifier-Guided Access Control Policy SynthesisLLM agents and tool use → Agent threats, guardrails and governance
Do Medical Vision Language Models Actually See? A Counterfactual Grounding Framework and Hard-Negative Contrastive Training for Visually-Reliant Medical VLMsMultimodal LLMs / vision-language models → Medical vision-language models
GEPARD - Generative, Prosody-aware, Autoregressive text-to-speech model for Realtime Dialogue
Distilled Rapid Embedding Transfer (DRET): Parameter-Efficient Biomedical Domain Adaptation via Priority-Based Embedding Transfer
C3-UniMM: Causal Cycle-Consistent Unified Multimodal Modeling via Super Alignment and Shared Decoding Space
Selection Bias Correction in Retail Intelligence
Position: AI Leaderboards Are Underserving the Global South: A Case Study from India
Regulatory Approval Is Not Enough: Gaps in Trustworthy AI Reporting in FDA-Cleared Medical Devices
Interaction Readiness: A Framework for Building and Evaluating AI Agents in Human Roles
Measuring Curriculum-Labor Market Alignment at the Scale of a Program Portfolio
Cloud-Native Evaluation-as-a-Service: A Microservices Architecture for Scalable AI Monitoring with Conformal Guarantees
Memory Merge DQN: Sensitivity Weighted Target Updates for Stable Value Learning
Data-Driven Forward and Inverse Modeling of V-Beam Thermal Sensors
SHIFT: Survival Prediction from Incomplete and Heterogeneous Genomic Data
Uncertainty-gated selection for block-sparse attention
Omni-Sleep: A Sleep Foundation Model via Hierarchical Contrastive Learning of CNS-ANS Dynamics
ReCoLoRA: Spectrum-Aware Recursive Consolidation for Continual LLM Fine-Tuning
LLT: Local Linear Transformer for PDE Operator Learning
Who Gets Missed in the Tail? Thresholded Subgroup Underdiagnosis in Long-Tailed Chest X-ray Classification
Towards the Explainability of Temporal Graph Networks via Memory Backtracking and Topological Attribution
Trajectory Inference of Human Aging from Cross-Sectional DNA Methylation Data
BitFair: A 12-nm Bit-Serial CNN Accelerator with Learnable Early Termination and Adaptive Bit Ordering for Ultra-Low-Power XR Vision
PreSIST: Vision-Language-Informed Object Persistence Prediction in Open-World Scenes
A Policy Decomposition Framework for Dynamic Order Fulfillment Operations
What is Left for Us? Second Scholarship Against the Degradation of Research by AI
SiamJEPA: On the Role of Siamese Student Encoders in JEPA
OmniOpt: Taxonomy, Geometry, and Benchmarking of Modern Optimizers
Efficient Discovery of Conditional Dependencies with Desbordante
CrossHallu: Do Hallucination Signals Generalize Across Languages and Domains in Large Language Model's Internals?
A Unified Algebraic Framework for Classification Performance Evaluation
Paired Uterine Whole-Slide Images and Pathology Reports for Multimodal Computational Pathology
Finite Reliability Representations: Noise-Calibrated Belief-Space Covers for Reliable Decision-Making
SAGE: Synchronized Action-Gaze Recognition and Anticipation for Human Behavior Understanding
When Does Small Data Work? Accuracy and Efficiency Trade-offs Between Tabular Foundation Models and Conventional Methods for Crowd-State Classification at Hajj and Umrah
Separating Representation from Reconstruction Enables Scalable Text Encoders
Explainable AI for Screening Abuse-Related Trauma in Bangladeshi Children: A Training-Free Multimodal Framework Evaluated on Noise-Aware Synthetic Data
Significance-First Splitting: Aligning Treatment Heterogeneity Detection with Honest Estimation
Directional Curvature from Armijo Backtracking: A Low-Cost Sharpness Probe and a Calibration-Free Learning-Rate Safeguard for Adam
One Image, No Tokens: A Controlled Study of Glyph-Based Chinese Language Modeling
Knowing When to Stop: Predicting Execution-Consistency Convergence in Text-to-SQL
InSpace: Structure-Aware 3D Indoor Scene Generation from a Single 360deg Image
NouveauVoice: Generating Novel Pseudo Speakers for Voice Anonymization
BanglaMemeEvidence: A Multimodal Benchmark Dataset for Explanatory Evidence Detection in Bengali Memes
Scalable Semantic Steering of Embedding Projections
TRACER: Early Failure Detection for Task-Oriented Dialogue
MANCE: Manifold Aware Concept Erasure

All 188