Papers

New on arXiv

Sat 7 Feb 2026

213 papers

Developing an AI Assistant for Knowledge Management and Workforce Training in State DOTsRetrieval-augmented generation → Agentic RAG, agents and memory
Benchmarking Legal RAG: The Promise and Limits of AI Statutory SurveysRetrieval-augmented generation → Legal and financial RAG
AngelSlim: A more accessible, comprehensive, and efficient toolkit for large model compressionEfficient LLM inference and serving → KV cache optimization
Agents in the Wild: Safety, Society, and the Illusion of Sociality on MoltbookLLM safety and security → Prompt injection and agent security
AgentTrace: A Structured Logging Framework for Agent System ObservabilityLLM agents and tool use → Agent threats, guardrails and governance
ParisKV: Fast and Drift-Robust KV-Cache Retrieval for Long-Context LLMsEfficient LLM inference and serving → KV cache optimization
Efficient Planning in Reinforcement Learning via Model IntrospectionDeep reinforcement learning methods → Deep RL algorithms and representations
Analyzing and Guiding Zero-Shot Posterior Sampling in Diffusion ModelsDiffusion models → Inverse problems and image restoration
On Sequence-to-Sequence Models for Automated Log ParsingState-space sequence models (S4, Mamba) → Core SSM architectures for language
Vision and Language: Novel Representations and Artificial intelligence for Driving Scene Safety Assessment and Autonomous Vehicle PlanningMultimodal LLMs / vision-language models → Driving VLMs and open-vocabulary detection +1
Debugging code world modelsWorld models → World models for language agents
Agent-Fence: Mapping Security Vulnerabilities Across Deep Research AgentsLLM agents and tool use → Agent threats, guardrails and governance
Efficient Table Retrieval and Understanding with Multimodal Large Language ModelsMultimodal LLMs / vision-language models → VLM benchmarks and visual reasoning
HAIF: A Human-AI Integration Framework for Hybrid Team OperationsLLM agents and tool use → Software engineering coding agents
Letting Tutor Personas Speak Up for LLMs: Learning Steering Vectors from Dialogue via Preference OptimizationLLM alignment / RLHF / preference optimisation → Process reward models for reasoning +1
SERE: Similarity-based Expert Re-routing for Efficient Batch Decoding in MoE ModelsMixture-of-experts → Efficient MoE training and inference systems
Fine-R1: Make Multi-modal LLMs Excel in Fine-Grained Visual Recognition by Chain-of-Thought ReasoningLLM reasoning → Multimodal reasoning +1
Learning to Self-Verify Makes Language Models Better ReasonersLLM reasoning → Test-time compute and efficient thinking
ViCA: Efficient Multimodal LLMs with Vision-Only Cross-AttentionMultimodal LLMs / vision-language models → Visual token pruning and efficiency
SIGMA: Selective-Interleaved Generation with Multi-Attribute TokensDiffusion models → Image generation, text-to-image and editing
VERIFY-RL: Verifiable Recursive Decomposition for Reinforcement Learning in Mathematical ReasoningLLM reasoning → Mathematical reasoning
VISOR: VIsual Spatial Object Reasoning for Language-driven Object NavigationLLM reasoning → Spatial and embodied reasoning
Capturing the Topological Phase Transition and Thermodynamics of the 2D XY Model via Manifold-Aware Score-Based Generative ModelingDiffusion models → Diffusion foundations and sampling
Improving Variable-Length Generation in Diffusion Language Models via Length RegularizationDiffusion models → Discrete and text diffusion
Differentiate-and-Inject: Enhancing VLAs via Functional Differentiation Induced by In-Parameter Structural ReasoningVision-language-action models → Reasoning and chain-of-thought in VLA
LLM-Guided Diagnostic Evidence Alignment for Medical Vision-Language Pretraining under Limited PairingMultimodal LLMs / vision-language models → Medical vision-language models
Joint Reward Modeling: Internalizing Chain-of-Thought for Efficient Visual Reward ModelsLLM reasoning → Multimodal reasoning +1
MedVerse: Efficient and Reliable Medical Reasoning via DAG-Structured Parallel ExecutionLLM reasoning → Clinical and biomedical applications
IGMiRAG: Intuition-Guided Retrieval-Augmented Generation with Adaptive Mining of In-Depth MemoryRetrieval-augmented generation → Knowledge-graph and structured-knowledge RAG
SPECA: Specification-to-Checklist Agentic Auditing for Multi-Implementation Systems -- A Case Study on Ethereum ClientsLLM agents and tool use → Agent threats, guardrails and governance
IM-Animation: An Implicit Motion Representation for Identity-decoupled Character AnimationDiffusion models → Video generation
From Native Memes to Global Moderation: Cross-Cultural Evaluation of Vision-Language Models for Hateful Meme DetectionMultimodal LLMs / vision-language models → Social bias and fairness in VLMs
CoMI-IRL: Contrastive Multi-Intention Inverse Reinforcement LearningOffline RL and imitation learning from data → Inverse reinforcement learning +1
GraphAgents: Knowledge Graph-Guided Agentic AI for Cross-Domain Materials DesignLLM agents and tool use → Medical and scientific domain agents
Are Reasoning LLMs Robust to Interventions on Their Chain-of-Thought?LLM reasoning → Mathematical reasoning
SpatialReward: Bridging the Perception Gap in Online RL for Image Editing via Explicit Spatial ReasoningLLM reasoning → Spatial and embodied reasoning
DLLM Agent: See Farther, Run FasterLLM agents and tool use → General LLM agent frameworks and benchmarks
Proximal Action Replacement for Behavior Cloning Actor-Critic in Offline Reinforcement LearningOffline RL and imitation learning from data → Imitation learning methods
Secure Code Generation via Online Reinforcement Learning with Vulnerability Reward ModelLLM alignment / RLHF / preference optimisation → Safety alignment and jailbreak attacks
Nonparametric Bayesian Optimization for General RewardsLLM alignment / RLHF / preference optimisation → Reward modeling and RLHF
Progressive Multi-Agent Reasoning for Biological Perturbation PredictionLLM reasoning → Clinical and biomedical applications +1
VGAS: Value-Guided Action-Chunk Selection for Few-Shot Vision-Language-Action AdaptationLLM reasoning → Spatial and embodied reasoning +1
AgentSys: Secure and Dynamic LLM Agents Through Explicit Hierarchical Memory ManagementLLM agents and tool use → Agent threats, guardrails and governance +1
Scout Before You Attend: Sketch-and-Walk Sparse Attention for Efficient LLM InferenceEfficient LLM inference and serving → KV cache optimization
NAAMSE: Framework for Evolutionary Security Evaluation of AgentsLLM agents and tool use → Agent threats, guardrails and governance +1
Trace-Focused Diffusion Policy for Multi-Modal Action Disambiguation in Long-Horizon Robotic ManipulationDiffusion models → Diffusion and flow policies for robotics
When the Model Said 'No Comment', We Knew Helpfulness Was Dead, Honesty Was Alive, and Safety Was TerrifiedMixture-of-experts → MoE large language models
Aegis: Towards Governance, Integrity, and Security of AI Voice AgentsLLM agents and tool use → Agent threats, guardrails and governance +1
Dichotomy of Feature Learning and Unlearning: Fast-Slow Analysis on Neural Networks with Stochastic Gradient DescentLLM safety and security → Machine unlearning
ViHERMES: A Graph-Grounded Multihop Question Answering Benchmark and System for Vietnamese Healthcare RegulationsLLM reasoning → Knowledge- and tool-augmented reasoning
W&D:Scaling Parallel Tool Calling for Efficient Deep Research AgentsLLM agents and tool use → Web, search, and research agents
Controllable Value Alignment in Large Language Models through Neuron-Level EditingLLM alignment / RLHF / preference optimisation → Human values and AI feedback alignment
Seeing Roads Through Words: A Language-Guided Framework for RGB-T Driving Scene SegmentationEnd-to-end autonomous driving → VLMs for autonomous driving
Revisiting Robustness for LLM Safety Alignment via Selective Geometry ControlLLM alignment / RLHF / preference optimisation → Safety alignment and jailbreak attacks +1
RAPiD: Reward-Guided Consistency Distillation of Diffusion Planners for Real-Time Autonomous DrivingEnd-to-end autonomous driving → Imitation-learning end-to-end planners
Meta-Reinforcement Learning for Robust and Non-greedy Control Barrier Functions in Spacecraft Proximity OperationsDeep reinforcement learning methods → RL for robot navigation and UAV control
High Fidelity Textual User Representation over Heterogeneous Sources via Reinforcement LearningDeep reinforcement learning methods → RL training of language and agent models
Action-to-Action Flow MatchingDiffusion models → Diffusion and flow policies for robotics
LUCID-SAE: Learning Unified Vision-Language Sparse Codes for Interpretable Concept DiscoveryMechanistic interpretability → Interpreting vision and multimodal models
Principled Synthetic Data Enables the First Scaling Laws for LLMs in RecommendationSynthetic data and self-improvement for LLMs → LLM synthetic data generation
Progressive Searching for Retrieval in RAGRetrieval-augmented generation → RAG pipeline design and evaluation
Patch-to-PoC: A Systematic Study of Agentic LLM Systems for Linux Kernel N-Day ReproductionLLM agents and tool use → Agent threats, guardrails and governance
Mapping the Design Space of User Experience for Computer Use AgentsLLM agents and tool use → GUI and mobile agents
Cross-View World ModelsWorld models → Video-generation world models and simulators
Steer2Adapt: Dynamically Composing Steering Vectors Elicits Efficient Adaptation of LLMsMechanistic interpretability → Activation steering methods
A Rational Account of Categorization Based on Information Theory
A Holistic Framework for Automated Configuration Recommendation for Cloud Service Monitoring
Knowledge Graph and Hypergraph Transformers with Repository-Attention and Journey-Based Role Transport
HumanLM: Simulating Users with State Alignment Beats Response Imitation
From Exact Hits to Close Enough: Semantic Caching for LLM Embeddings
How LLMs Cite and Why It Matters: A Cross-Model Audit of Reference Fabrication in AI-Assisted Academic Writing and Methods to Detect Phantom Citations
Patient-Centered, Graph-Augmented Artificial Intelligence-Enabled Passive Surveillance for Early Stroke Risk Detection in High-Risk Individuals
Urban Vibrancy Embedding and Application on Traffic Prediction
Deep Neural Network Architectures for Electrocardiogram Classification: A Comprehensive Evaluation
Surrogate Modeling for Neutron Transport: A Neural Operator Approach
Benchmarking Anomaly Detection Across Heterogeneous Cloud Telemetry Datasets
COOPERTRIM: Adaptive Data Selection for Uncertainty-Aware Cooperative Perception
Explanatory Interactive Machine Learning for Bias Mitigation in Visual Gender Classification
Reverse-Engineering Model Editing on Language Models
SVD-Preconditioned Gradient Descent Method for Solving Nonlinear Least Squares Problems
Efficient Adaptive Data Analysis over Dense Distributions
The Laplacian Keyboard: Beyond the Linear Span
Do We Need Adam? Surprisingly Strong and Sparse Reinforcement Learning with SGD in LLMs
All-Optical Segmentation via Diffractive Neural Networks for Autonomous Driving
Towards Robust Scaling Laws for Optimizers
On Generation in Metric Spaces
Quantifying Explanation Quality in Graph Neural Networks using Out-of-Distribution Generalization
Dense Feature Learning via Linear Structure Preservation in Medical Data
A hybrid Kolmogorov-Arnold network for medical image segmentation
On the Infinite Width and Depth Limits of Predictive Coding Networks
EventCast: Hybrid Demand Forecasting in E-Commerce with LLM-Based Event Knowledge
Semantic-Deviation-Anchored Multi-Branch Fusion for Unsupervised Anomaly Detection and Localization in Unstructured Conveyor-Belt Coal Scenes
Process-of-Thought Reasoning for Videos
Mapping Drivers of Greenness: Spatial Variable Selection for MODIS Vegetation Indices
Spectral Gating Networks
ElliCE: Efficient and Provably Robust Algorithmic Recourse via the Rashomon Sets
Blind to the Human Touch: Overlap Bias in LLM-Based Summary Evaluation
Federated Learning with Profile Mapping under Distribution Shifts and Drifts
Surprisal-Guided Selection: Compute-Optimal Test-Time Strategies for Execution-Grounded Code Generation
Looking and Listening Inside and Outside: Multimodal Artificial Intelligence Systems for Driver Safety Assessment and Intelligent Vehicle Decision-Making

All 213