Papers

New on arXiv

Sun 8 Mar 2026

232 papers

GraphBit: A Graph-based Agentic Framework for Non-Linear Agent OrchestrationLLM agents and tool use → Software engineering coding agents
Evaluating LLMs as Human Surrogates in Controlled ExperimentsSynthetic data and self-improvement for LLMs → Self-rewarding alignment and self-training
MDKeyChunker: Single-Call LLM Enrichment with Rolling Keys and Key-Based Restructuring for High-Accuracy RAGRetrieval-augmented generation → Multimodal RAG
Beyond Hard Constraints: Budget-Conditioned Reachability For Safe Offline Reinforcement LearningOffline RL and imitation learning from data → Safe offline RL
Emergency Lane-Change Simulation: A Behavioral Guidance Approach for Risky Scenario GenerationDeep reinforcement learning methods → RL for autonomous driving and traffic
InfoMamba: An Attention-Free Hybrid Mamba-Transformer ModelState-space sequence models (S4, Mamba) → Core SSM architectures for language
Quine: Realizing LLM Agents as Native POSIX ProcessesLLM agents and tool use → Software engineering coding agents
Optimizing LLM Annotation of Classroom Discourse through Multi-Agent OrchestrationLLM agents and tool use → Educational agents and math problem solving
Local Precise Refinement: A Dual-Gated Mixture-of-Experts for Enhancing Foundation Model Generalization against Spectral ShiftsMixture-of-experts → Vision mixture-of-experts
Intentional Deception as Controllable Capability in LLM AgentsLLM agents and tool use → Social simulation and multi-agent debate
Relating Reinforcement Learning to Dynamic Programming-Based PlanningDeep reinforcement learning methods → Deep RL algorithms and representations
Training-free Temporal Object Tracking in Surgical VideosDiffusion models → Medical imaging and segmentation
DistillGuard: Evaluating Defenses Against LLM Knowledge DistillationLLM safety and security → Jailbreaking +1
Benchmarking Large Language Models for Quebec Insurance: From Closed-Book to Retrieval-Augmented GenerationRetrieval-augmented generation → Legal and financial RAG
Reasoning Knowledge-Gap in Drone Planning via LLM-based Active ElicitationLLM reasoning → Spatial and embodied reasoning
HybridStitch: Pixel and Timestep Level Model Stitching for Diffusion AccelerationDiffusion models → Image generation, text-to-image and editing
Preference-Conditioned Reinforcement Learning for Space-Time Efficient Online 3D Bin PackingDeep reinforcement learning methods → RL for robot navigation and UAV control
MWM: Mobile World Models for Action-Conditioned Consistent PredictionWorld models → World models for navigation
Toward Global Intent Inference for Human Motion by Inverse Reinforcement LearningOffline RL and imitation learning from data → Inverse reinforcement learning +1
OrdinalBench: A Benchmark Dataset for Diagnosing Generalization Limits in Ordinal Number Understanding of Vision-Language ModelsMultimodal LLMs / vision-language models → VLM benchmarks and visual reasoning
ProgAgent:A Continual RL Agent with Progress-Aware RewardsLLM alignment / RLHF / preference optimisation → Reward modeling and RLHF +1
Scaling Data Difficulty: Improving Coding Models via Reinforcement Learning on Fresh and Challenging ProblemsDeep reinforcement learning methods → RL training of language and agent models
Breaking Training Bottlenecks: Effective and Stable Reinforcement Learning for Coding ModelsDeep reinforcement learning methods → RL training of language and agent models
Geometric Knowledge-Assisted Federated Dual Knowledge Distillation Approach Towards Remote Sensing Satellite ImageryKnowledge distillation → Detection and dense prediction distillation
MedQ-Deg: A Multidimensional Benchmark for Evaluating MLLMs Across Medical Image Quality DegradationsMultimodal LLMs / vision-language models → Medical vision-language models
DECADE: A Temporally-Consistent Unsupervised Diffusion Model for Enhanced Rb-82 Dynamic Cardiac PET Image DenoisingDiffusion models → Medical imaging and segmentation
3ViewSense: Spatial and Mental Perspective Reasoning from Orthographic Views in Vision-Language ModelsLLM reasoning → Spatial and embodied reasoning +1
Large Language Model for Discrete Optimization Problems: Evaluation and Step-by-step ReasoningLLM reasoning → Mathematical reasoning
A Novel Multi-Agent Architecture to Reduce Hallucinations of Large Language Models in Multi-Step Structural ModelingLLM agents and tool use → Multimodal and embodied agents
Verifiable Reasoning for LLM-based Generative RecommendationLLM reasoning → LLM reasoning for recommendation
TDM-R1: Reinforcing Few-Step Diffusion Models with Non-Differentiable RewardDiffusion models → Diffusion and flow policies for robotics +2
Global Convergence of Average Reward Constrained MDPs with Neural Critic and General Policy ParameterizationDeep reinforcement learning methods → RL theory and sample complexity
Learning Context-Adaptive Motion Priors for Masked Motion Diffusion Models with Efficient Kinematic Attention AggregationDiffusion models → Video generation
FrameVGGT: Coherence-Preserving Memory for Bounded Streaming GeometryEfficient LLM inference and serving → Token pruning and merging for ViTs
UniUncer: Unified Dynamic Static Uncertainty for End to End DrivingEnd-to-end autonomous driving → End-to-end system architecture
Scalable Training of Mixture-of-Experts Models with Megatron CoreMixture-of-experts → Efficient MoE training and inference systems
Memory for Autonomous LLM Agents:Mechanisms, Evaluation, and Emerging FrontiersLLM agents and tool use → Agent memory systems
Scaling Test-Time Robustness of Vision-Language Models via Self-Critical Inference FrameworkMultimodal LLMs / vision-language models → VLM benchmarks and visual reasoning
AtomicVLA: Unlocking the Potential of Atomic Skill Learning in RobotsMixture-of-experts → MoE in reinforcement learning and robotics
TempoFit: Plug-and-Play Layer-Wise Temporal KV Memory for Long-Horizon Vision-Language-Action ManipulationVision-language-action models → Reasoning and chain-of-thought in VLA
Helix: Evolutionary Reinforcement Learning for Open-Ended Scientific Problem SolvingLLM reasoning → Reinforcement learning for reasoning +1
Accelerating Diffusion Models for Generative AI Applications with Silicon PhotonicsDiffusion models → Diffusion foundations and sampling
Overthinking Causes Hallucination: Tracing Confounder Propagation in Vision Language ModelsLLM reasoning → Hallucination detection and mitigation +1
KohakuRAG: A simple RAG framework with hierarchical document indexingRetrieval-augmented generation → RAG pipeline design and evaluation
SCA: Segment-Wise CoT Compression with Answer AlignmentLLM reasoning → Test-time compute and efficient thinking
Models as Lego Builders: Assembling Malice from Benign Blocks via Semantic BlueprintsMultimodal LLMs / vision-language models → MLLM jailbreaks and adversarial robustness
Approximate Imitation Learning for Event-based Quadrotor Flight in Cluttered EnvironmentsOffline RL and imitation learning from data → Imitation learning methods
TS-MLLM: A Multi-Modal Large Language Model-based Framework for Industrial Time-Series Big Data AnalysisMultimodal LLMs / vision-language models → Chart understanding and anomaly detection
Constraints Matrix Diffusion based Generative Neural Solver for Vehicle Routing ProblemsDiffusion models → Diffusion foundations and sampling
Revisiting the LiRA Membership Inference Attack Under Realistic AssumptionsLLM safety and security → Membership inference attacks
Brain-WM: Brain Glioblastoma World ModelWorld models → Medical and surgical world models
AgentRaft: Automated Detection of Data Over-Exposure in LLM AgentsLLM agents and tool use → Agent threats, guardrails and governance
ReconDrive: Fast Feed-Forward 4D Gaussian Splatting for Autonomous Driving Scene ReconstructionEnd-to-end autonomous driving → Generative simulation for driving
COOL-MC: Verifying and Explaining RL Policies for Multi-bridge Network MaintenanceDeep reinforcement learning methods → RL for energy, networks and operations
DreamSAC: Learning Hamiltonian World Models via Symmetry ExplorationWorld models → Latent predictive world models
CONSTANT: Towards High-Quality One-Shot Handwriting Generation with Patch Contrastive Enhancement and Style-Aware QuantizationDiffusion models → Image generation, text-to-image and editing
MAWARITH: A Dataset and Benchmark for Legal Inheritance Reasoning with LLMsLLM reasoning → Reasoning for agents, planning and code
ICLR: In-Context Imitation Learning with Visual ReasoningLLM reasoning → Spatial and embodied reasoning +1
TableMind++: An Uncertainty-Aware Programmatic Agent for Tool-Augmented Table ReasoningLLM reasoning → Knowledge- and tool-augmented reasoning +1
Reinforcement learning-based dynamic cleaning scheduling framework for solar energy systemDeep reinforcement learning methods → RL for energy, networks and operations
EvolveReason: Self-Evolving Reasoning Paradigm for Explainable Deepfake Facial Image IdentificationSynthetic data and self-improvement for LLMs → Self-improving MLLMs and synthetic data
A Unified View of Score-Based and Drifting ModelsDiffusion models → Diffusion foundations and sampling
High-Fidelity Medical Shape Generation via Skeletal Latent DiffusionDiffusion models → Medical imaging and segmentation
From Thinker to Society: Security in Hierarchical Autonomy Evolution of AI AgentsLLM agents and tool use → Agent threats, guardrails and governance
DocCogito: Aligning Layout Cognition and Step-Level Grounded Reasoning for Document UnderstandingLLM reasoning → Multimodal reasoning +1
A Comparative analysis of Layer-wise Representational Capacity in AR and Diffusion LLMsDiffusion models → Discrete and text diffusion
Give Them an Inch and They Will Take a Mile:Understanding and Measuring Caller Identity Confusion in MCP-Based AI SystemsLLM agents and tool use → Agent threats, guardrails and governance
Contact-Guided 3D Genome Structure Generation of E. coli via Diffusion TransformersDiffusion models → Flow matching, time series and science
Selective Transfer Learning of Cross-Modality Distillation for Monocular 3D Object DetectionKnowledge distillation → Detection and dense prediction distillation
Backdoor4Good: Benchmarking Beneficial Uses of Backdoors in LLMsLLM safety and security → Backdoor and data poisoning attacks
Few Tokens, Big Leverage: Preserving Safety Alignment by Constraining Safety Tokens during Fine-tuningLLM alignment / RLHF / preference optimisation → Safety alignment and jailbreak attacks +1
HLER: Human-in-the-Loop Economic Research via Multi-Agent Pipelines for Empirical DiscoveryLLM agents and tool use → Financial trading and forecasting agents
Med-Evo: Test-time Self-evolution for Medical Multimodal Large Language ModelsMultimodal LLMs / vision-language models → Medical vision-language models
Generalization in Online Reinforcement Learning for Mobile AgentsLLM agents and tool use → GUI and mobile agents +1
Disentangled Textual Priors for Diffusion-based Image Super-ResolutionDiffusion models → Inverse problems and image restoration
DualSpec: Accelerating Deep Research Agents via Dual-Process Action SpeculationLLM agents and tool use → Web, search, and research agents
Adaptive Capacity Allocation for Vision Language Action Fine-tuningVision-language-action models → Generalist VLA models and benchmarks
AQuA: Toward Strategic Response Generation for Ambiguous Visual QuestionsMultimodal LLMs / vision-language models → VLM benchmarks and visual reasoning
Benchmark for Assessing Olfactory Perception of Large Language Models
Filter-then-Weight: Online Data Selection and Reweighting for LLM Fine-Tuning
Power Couple? AI Growth and Renewable Energy Investment
Not All Pretraining are Created Equal: Threshold Tuning and Class Weighting for Imbalanced Polarization Tasks in Low-Resource Settings
Generating Hierarchical JSON Representations of Scientific Sentences Using LLMs
Writing literature reviews with AI: principles, hurdles and some lessons learned
Generalized Stock Price Prediction for Multiple Stocks Combined with News Fusion
Beam-aware Kernelized Contextual Bandits for User Association and Beamforming in mmWave Vehicular Networks
CDEoH: Category-Driven Automatic Algorithm Design With Large Language Models
Engineering Verifiable Modularity in Transformers via Per-Layer Supervision
What on Earth is AlphaEarth? Hierarchical structure and functional interpretability for global land cover
Machine Learning Based Identification of Solvents from Post-Desiccation Patterns
Did You Check the Right Pocket? Cost-Sensitive Store Routing for Memory-Augmented Agents
Attribution-Guided Model Rectification of Unreliable Neural Network Behaviors
AgriPath: A Systematic Exploration of Architectural Trade-offs for Crop Disease Classification
From Garbage to Gold: A Data-Architectural Theory of Predictive Robustness
Context-Enriched Natural Language Descriptions of Vessel Trajectories
Toward Epistemic Stability: Engineering Consistent Procedures for Industrial LLM Hallucination Reduction
Gated Adaptation for Continual Learning in Human Activity Recognition
Safety Under Scaffolding: How Evaluation Conditions Shape Measured Safety
No Certificate for Alignment: Two Independent Impossibilities and the Pareto Frontier of Achievable Safety Guarantees
Clear, Compelling Arguments: Rethinking the Foundations of Frontier AI Safety Cases

All 232