Liang Lin

Sun Yat-sen University

収録論文 45本 ・ フィジカルAI/ロボット学習

※arXiv著者名で収集。同姓同名の別人の論文が含まれる場合があります。

論文

• WorldSimProbe: 身体的操作のための行動条件付きワールドモデルにおけるシミュレータ忠実度の診断
  シミュレータ評価2026/8/10

  行動条件付きワールドモデル（ACWM）が物理シミュレータとして忠実に機能するかを検証するための評価手法WorldSimProbeを提案し、複数のベンチマークで既存モデルの系統的な欠陥を明らかにした。

• CrossTracer: VLAモデルの推論とトレース残差適応によるクロスエンボディメントナビゲーション
  ナビゲーション2026/8/7

  ロボットの種類に応じた移動制約を考慮したナビゲーション経路を生成する階層的フレームワークを提案。VLAモデルによる初期経路生成と、エンボディメント条件付きの残差補正を組み合わせ、汎用ベースラインを上回る性能を達成した。

• WorldSimProbe: Diagnosing Simulator Faithfulness in Action-Conditioned World Models for Embodied Manipulation
  2026/8/1
• CrossTracer: Cross-Embodiment Navigation via VLA Model Reasoning and Trace Residuals Adapting
  2026/8/1
• PhyAgentOS: A Self-Evolving Operating System for Embodied Agents with Decoupled Cognitive Planning and Physical Execution
  2026/7/1
• Joint On-and-Off Policy Learning for Vision-and-Language Navigation
  2026/7/1
• Bridge-WA: Predicting Where and How the World Changes for Robotic Action
  2026/7/1
• NavIsaacLab: Generating Realistic Crowd via Parallel Robot Learning for Benchmarking Human-aware Navigation
  2026/6/1
• PACE: Phase-Aware Chunk Execution for Robot Policies with Action Chunking
  2026/6/1
• JoyAI-Sim: A Simulation-Enabled Interconversion Toolchain for the Embodied Data Pyramid
  2026/6/1
• In-Context Model Predictive Generation: Open-Vocabulary Motion Synthesis from Language Models to Physics
  2026/6/1
• RoVLA: Multi-Consistency Constraints for Robust Vision-Language-Action Models
  2026/5/1
• RePO-VLA: Recovery-Driven Policy Optimization for Vision-Language-Action Models
  2026/5/1
• SkiP: When to Skip and When to Refine for Efficient Robot Manipulation
  2026/5/1
• Uncovering Linguistic Fragility in Vision-Language-Action Models via Diversity-Aware Red Teaming
  2026/4/1
• Robotic Manipulation is Vision-to-Geometry Mapping: Vision-Geometry Backbones over Language and Video Models
  2026/4/1
• EgoLive: A Large-Scale Egocentric Dataset from Real-World Human Tasks
  2026/4/1
• JoyAI-RA 0.1: A Foundation Model for Robotic Autonomy
  2026/4/1
• Referring-Aware Visuomotor Policy Learning for Closed-Loop Manipulation
  2026/4/1
• AtomicVLA: Unlocking the Potential of Atomic Skill Learning in Robots
  2026/3/1
• Learning Physics from Pretrained Video Models: A Multimodal Continuous and Sequential World Interaction Models for Robotic Manipulation
  2026/3/1
• TAG: Target-Agnostic Guidance for Stable Object-Centric Inference in Vision-Language-Action Models
  2026/3/1
• DDP-WM: Disentangled Dynamics Prediction for Efficient World Models
  2026/2/1
• RADAR: Benchmarking Vision-Language-Action Generalization via Real-World Dynamics, Spatial-Physical Intelligence, and Autonomous Evaluation
  2026/2/1
• Stable Language Guidance for Vision-Language-Action Models
  2026/1/1
• AdaMorph: Unified Motion Retargeting via Embodiment-Aware Adaptive Transformers
  2026/1/1
• VLA Models Are More Generalizable Than You Think: Revisiting Physical and Spatial Modeling
  2025/12/2
• VLA Models Are More Generalizable Than You Think: Revisiting Physical and Spatial Modeling
  2025/12/1
• E0: Enhancing Generalization and Fine-Grained Control in VLA Models via Tweedie Discrete Diffusion
  2025/11/26
• E0: Enhancing Generalization and Fine-Grained Control in VLA Models via Tweedie Discrete Diffusion
  2025/11/1
• GraspView: Active Perception Scoring and Best-View Optimization for Robotic Grasping in Cluttered Environments
  2025/11/1
• RoVer: Robot Reward Model as Test-Time Verifier for Vision-Language-Action Model
  2025/10/1
• Embodied Arena: A Comprehensive, Unified, and Evolving Evaluation Platform for Embodied AI
  2025/9/1
• Learning to See and Act: Task-Aware Virtual View Exploration for Robotic Manipulation
  2025/8/1
• AutoLayout: Closed-Loop Layout Synthesis via Slow-Fast Collaborative Reasoning
  2025/7/1
• RoboPearls: Editable Video Simulation for Robot Manipulation
  2025/6/1
• RoBridge: A Hierarchical Architecture Bridging Cognition and Execution for General Robotic Manipulation
  2025/5/1
• InfiniteWorld: A Unified Scalable Simulation Framework for General Visual-Language Robot Interaction
  2024/12/1
• All Robots in One: A New Standard and Unified Dataset for Versatile, General-Purpose Embodied Agents
  2024/8/1
• Aligning Cyber Space with Physical World: A Comprehensive Survey on Embodied AI
  2024/7/1
• Towards Deviation-Robust Agent Navigation via Perturbation-Aware Contrastive Learning
  2024/3/1
• Reward-Adaptive Reinforcement Learning: Dynamic Policy Gradient Optimization for Bipedal Locomotion
  2021/7/1
• Continuous Transition: Improving Sample Efficiency for Continuous Control Problems via MixUp
  2020/11/1
• NADPEx: An on-policy temporally consistent exploration method for deep reinforcement learning
  2018/12/1
• Neural Task Planning with And-Or Graph Representations
  2018/8/1