Papers

New on arXiv

Sat 22 Nov 2025

195 papers

WebSTAR: Scalable Data Synthesis for Computer Use Agents with Step-Level FilteringLLM agents and tool use → GUI and mobile agents +1
Scaling Competence, Shrinking Reasoning: Cognitive Signatures in Language Model LearningLLM reasoning → Test-time compute and efficient thinking
Quality analysis and evaluation prediction of RAG retrieval based on machine learning algorithmsRetrieval-augmented generation → RAG pipeline design and evaluation
Exploiting the Experts: Unauthorized Compression in MoE-LLMsMixture-of-experts → Expert routing strategies
Building Browser Agents: Architecture, Security, and Practical SolutionsLLM agents and tool use → Agent threats, guardrails and governance
WavefrontDiffusion: Dynamic Decoding Schedule for Improved ReasoningLLM reasoning → Diffusion reasoning and test-time scaling +1
Enhancing Large Language Models for Automated Homework Assessment in Undergraduate Circuit AnalysisMechanistic interpretability → Circuits and mechanistic analysis
Deep Gaussian Process Proximal Policy OptimizationDeep reinforcement learning methods → RL theory and sample complexity
InfiniBench: Infinite Benchmarking for Visual Spatial Reasoning with Customizable Scene ComplexityLLM reasoning → Spatial and embodied reasoning +1
Agent-as-a-Graph: Knowledge Graph-Based Tool and Agent Retrieval for LLM Multi-Agent SystemsLLM agents and tool use → Web, search, and research agents
ARIAL: An Agentic Framework for Document VQA with Precise Answer LocalizationLLM agents and tool use → Multimodal and embodied agents
Accelerating Time Series Foundation Models with Speculative DecodingEfficient LLM inference and serving → Speculative decoding
Early Lung Cancer Diagnosis from Virtual Follow-up LDCT Generation via Correlational Autoencoder and Latent Flow MatchingDiffusion models → Medical imaging and segmentation
Rethinking Retrieval: From Traditional Retrieval Augmented Generation to Agentic and Non-Vector Reasoning Systems in the Financial Domain for Large Language ModelsLLM reasoning → Knowledge- and tool-augmented reasoning +2
Bringing Stability to Diffusion: Decomposing and Reducing Variance of Training Masked Diffusion ModelsDiffusion models → Discrete and text diffusion
UnfoldLDM: Degradation-Aware Unfolding with Iterative Latent Diffusion Priors for Blind Image RestorationDiffusion models → Inverse problems and image restoration
Observer-Actor: Active Vision Imitation Learning with Sparse-View Gaussian SplattingOffline RL and imitation learning from data → Robot manipulation imitation learning
Bias Is a Subspace, Not a Coordinate: A Geometric Rethinking of Post-hoc Debiasing in Vision-Language ModelsLLM reasoning → Multimodal reasoning +1
VCU-Bridge: Hierarchical Visual Connotation Understanding via Semantic BridgingMultimodal LLMs / vision-language models → VLM benchmarks and visual reasoning
PromptMoE: Generalizable Zero-Shot Anomaly Detection via Visually-Guided Prompt MixturesMixture-of-experts → Vision mixture-of-experts
ASTRA: Agentic Steerability and Risk Assessment FrameworkLLM agents and tool use → Agent threats, guardrails and governance +1
EchoVLA: Robotic Vision-Language-Action Model with Synergistic Declarative Memory for Mobile ManipulationVision-language-action models → Reasoning and chain-of-thought in VLA
Consolidating Diffusion-Generated Video Detection with Unified Multimodal Forgery LearningDiffusion models → Image generation, text-to-image and editing
A New Error Temporal Difference Algorithm for Deep Reinforcement Learning in Microgrid OptimizationDeep reinforcement learning methods → RL for energy, networks and operations
Continually Evolving Skill Knowledge in Vision Language Action ModelVision-language-action models → Reinforcement learning and post-training
The Alignment Paradox of Medical Large Language Models in Infertility Care: Decoupling Algorithmic Improvement from Clinical Decision-making QualityLLM reasoning → Clinical and biomedical applications +1
ActDistill: General Action-Guided Self-Derived Distillation for Efficient Vision-Language-Action ModelsVision-language-action models → Efficient VLA inference and adaptation
Reinforcement Learning for Portfolio Optimization with a Financial Goal and Defined Time HorizonsDeep reinforcement learning methods → Adversarial robustness and RL in finance
VK-Det: Visual Knowledge Guided Prototype Learning for Open-Vocabulary Aerial Object DetectionMultimodal LLMs / vision-language models → Embodied and 3D vision-language models
UltraFlux: Data-Model Co-Design for High-quality Native 4K Text-to-Image Generation across Diverse Aspect RatiosDiffusion models → Diffusion foundations and sampling
Curvature-Aware Safety Restoration In LLMs Fine-TuningLLM alignment / RLHF / preference optimisation → Safety alignment and jailbreak attacks +1
MASTEST: A LLM-Based Multi-Agent System For RESTful API TestsLLM agents and tool use → Software engineering coding agents
On a Reinforcement Learning Methodology for Epidemic Control, with application to COVID-19Deep reinforcement learning methods → RL for clinical treatment and surgery
Diverse Instance Generation via Diffusion Models for Enhanced Few-Shot Object Detection in Remote Sensing ImagesDiffusion models → Inverse problems and image restoration
RoadBench: Benchmarking MLLMs on Fine-Grained Spatial Understanding and Reasoning under Urban Road ScenariosLLM reasoning → Spatial and embodied reasoning +1
Reward Engineering for Spatial Epidemic Simulations: A Reinforcement Learning Platform for Individual Behavioral LearningDeep reinforcement learning methods → RL for clinical treatment and surgery
Privacy Auditing of Multi-domain Graph Pre-trained Model under Membership Inference AttacksLLM safety and security → Membership inference attacks
HyM-UNet: Synergizing Local Texture and Global Context via Hybrid CNN-Mamba Architecture for Medical Image SegmentationState-space sequence models (S4, Mamba) → Mamba for medical image analysis
Plan-X: Instruct Video Generation via Semantic PlanningDiffusion models → Video generation
Towards Effective, Stealthy, and Persistent Backdoor Attacks Targeting Graph Foundation ModelsLLM safety and security → Backdoor and data poisoning attacks
FeRA: Frequency-Energy Constrained Routing for Effective Diffusion Adaptation Fine-TuningDiffusion models → Diffusion foundations and sampling
Controllability Analysis of State Space-based Language ModelState-space sequence models (S4, Mamba) → Core SSM architectures for language
X-ReID: Multi-granularity Information Interaction for Video-Based Visible-Infrared Person Re-IdentificationMultimodal LLMs / vision-language models → Prompt learning and adaptation of CLIP
VITAL: Vision-Encoder-centered Pre-training for LMMs in Visual Quality AssessmentMultimodal LLMs / vision-language models → Image editing and quality assessment
Test-Time Temporal Sampling for Efficient MLLM Video UnderstandingMultimodal LLMs / vision-language models → Video understanding with MLLMs
SciEducator: Scientific Video Understanding and Educating via Deming-Cycle Multi-Agent SystemMultimodal LLMs / vision-language models → Video understanding with MLLMs +1
SPINE: Token-Selective Test-Time Reinforcement Learning with Entropy-Band RegularizationLLM reasoning → Reinforcement learning for reasoning +2
Alignment Faking - the Train -> Deploy Asymmetry: Through a Game-Theoretic Lens with Bayesian-Stackelberg EquilibriaLLM alignment / RLHF / preference optimisation → Human values and AI feedback alignment
MambaTAD: When State-Space Models Meet Long-Range Temporal Action DetectionState-space sequence models (S4, Mamba) → Video, motion and generative Mamba models
PA-FAS: Towards Interpretable and Generalizable Multimodal Face Anti-Spoofing via Path-Augmented Reinforcement LearningLLM reasoning → Multimodal reasoning +1
L2V-CoT: Cross-Modal Transfer of Chain-of-Thought Reasoning via Latent InterventionLLM reasoning → Multimodal reasoning +1
ChemVTS-Bench: Evaluating Visual-Textual-Symbolic Reasoning of Multimodal Large Language Models in ChemistryLLM reasoning → Clinical and biomedical applications +1
Principled Context Engineering for RAG: Statistical Guarantees via Conformal PredictionRetrieval-augmented generation → RAG robustness and adaptive retrieval
L1 Sample Flow for Efficient Visuomotor LearningDiffusion models → Diffusion and flow policies for robotics
MobileVLA-R1: Reinforcing Vision-Language-Action for Mobile RobotsVision-language-action models → Reasoning and chain-of-thought in VLA
When Better Teachers Don't Make Better Students: Revisiting Knowledge Distillation for CLIP Models in VQAMultimodal LLMs / vision-language models → MLLM training and visual instruction tuning +1
FastMMoE: Accelerating Multimodal Large Language Models through Dynamic Expert Activation and Routing-Aware Token PruningMultimodal LLMs / vision-language models → Visual token pruning and efficiency +2
ArticFlow: Generative Simulation of Articulated MechanismsDiffusion models → Diffusion and flow policies for robotics
Generative Adversarial Post-Training Mitigates Reward Hacking in Live Human-AI Music InteractionLLM alignment / RLHF / preference optimisation → Preference alignment for recommenders and RAG
Training Emergent Joint Associations: A Reinforcement Learning Approach to Creative Thinking in Language ModelsLLM reasoning → Reasoning benchmarks and evaluation
Beyond Jailbreak: Unveiling Risks in LLM Applications Arising from Blurred Capability BoundariesLLM safety and security → Jailbreaking
The Horcrux: Mechanistically Interpretable Task Decomposition for Detecting and Mitigating Reward Hacking in Embodied AI SystemsLLM alignment / RLHF / preference optimisation → Reward hacking and misspecification
QuickLAP: Quick Language-Action Preference Learning for Semi-Autonomous AgentsLLM agents and tool use → Reinforcement learning for LLM agents +1
Transformers with RL or SFT Provably Learn Sparse Boolean Functions, But DifferentlyLLM reasoning → Prompting and chain-of-thought reasoning
PPoGA: Predictive Plan-on-Graph with Action for Knowledge Graph Question Answering
Augmented Assembly: Object Recognition and Hand Tracking for Adaptive Assembly Instructions in Augmented Reality
Impact of Data-Oriented and Object-Oriented Design on Performance and Cache Utilization with Artificial Intelligence Algorithms in Multi-Threaded CPUs
ReVeal-MT: A Physics-Informed Neural Network for Multi-Transmitter Radio Environment Mapping
Partial multivariate transformer as a tool for cryptocurrencies time series prediction
DELTA: Language Diffusion-based EEG-to-Text Architecture
A Lightweight Approach to Detection of AI-Generated Texts Using Stylometric Features
The Human Brain as a Combinatorial Complex
LLMs-Powered Accurate Extraction, Querying and Intelligent Management of Literature derived 2D Materials Data
Federated Learning Framework for Scalable AI in Heterogeneous HPC and Cloud Environments
A Multi-Stage Deep Learning Framework with PKCP-MixUp Augmentation for Pediatric Liver Tumor Diagnosis Using Multi-Phase Contrast-Enhanced CT
FAST: Topology-Aware Frequency-Domain Distribution Matching for Coreset Selection
Tracking and Segmenting Anything in Any Modality
Pistachio: Towards Synthetic, Balanced, and Long-Form Video Anomaly Benchmarks
PrefixGPT: Prefix Adder Optimization by a Generative Pre-trained Transformer
Not Quite Anything: Overcoming SAMs Limitations for 3D Medical Imaging
Quantifying Modality Contributions via Disentangling Multimodal Representations
Towards a future space-based, highly scalable AI infrastructure system design
A Novel and Practical Universal Adversarial Perturbations against Deep Reinforcement Learning based Intrusion Detection Systems
Using MLIR Transform to Design Sliced Convolution Algorithm
Typing Reinvented: Towards Hands-Free Input via sEMG
Large-Scale Pre-training Enables Multimodal AI Differentiation of Radiation Necrosis from Brain Metastasis Progression on Routine MRI
ProHD: Projection-Based Hausdorff Distance Approximation
Generating Synthetic Human Blastocyst Images for In-Vitro Fertilization Blastocyst Grading
Improving Forecasts of Suicide Attempts for Patients with Little Data
Linear Algebraic Approaches to Neuroimaging Data Compression: A Comparative Analysis of Matrix and Tensor Decomposition Methods for High-Dimensional Medical Images
The Workflow as Medium: A Framework for Navigating Human-AI Co-Creation
MOMA-AC: A preference-driven actor-critic framework for continuous multi-objective multi-agent reinforcement learning
Bayesian Calibration of Engine-out NOx Models for Engine-to-Engine Transferability
Unified Spherical Frontend: Learning Rotation-Equivariant Representations of Spherical Images from Any Camera
EgoControl: Controllable Egocentric Video Generation via 3D Full-Body Poses
MEDIC: a network for monitoring data quality in collider experiments
BPMN to PDDL: Translating Business Workflows for AI Planning
Sparse Polyak with optimal thresholding operators for high-dimensional M-estimation
Towards a General Framework for HTN Modeling with LLMs
Nested Unfolding Network for Real-World Concealed Object Segmentation

All 195