Junbo Tan

Research Institute of Tsinghua University in Shenzhen

収録論文 14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

• 時間で検索し、周波数で修正する
  VLA2026/8/5

  凍結された視覚言語行動ポリシーの性能を、訓練なしのテスト時修正フレームワークで向上させる。成功軌道からの低周波残差を転送し、長期的な操作成功率を改善する。

• Retrieve in Time, Correct in Frequency
  2026/8/1
• FlowPRO: Reward-Free Reinforced Fine-Tuning of Flow-Matching VLAs via Proximalized Preference Optimization
  2026/6/1
• Learn Weightlessness: Imitate Non-Self-Stabilizing Motions on Humanoid Robot
  2026/4/1
• RPG: Robust Policy Gating for Smooth Multi-Skill Transitions in Humanoid Fighting
  2026/4/1
• PhotoAgent: A Robotic Photographer with Spatial and Aesthetic Understanding
  2026/3/1
• Choose What to Manipulate: Revealing Data Scaling Laws in Bounding-Box Guided Policies for Semantic Manipulation
  2026/2/1
• A Hybrid Force-Position Strategy for Shape Control of Deformable Linear Objects With Graph Attention Networks
  2025/8/1
• A Universal Vehicle-Trailer Navigation System with Neural Kinematics and Online Residual Learning
  2025/7/1
• Data-Driven MPC with Data Selection for Flexible Cable-Driven Robotic Arms
  2025/7/1
• FOSP: Fine-tuning Offline Safe Policy through World Models
  2024/7/1
• Offline Goal-Conditioned Reinforcement Learning for Safety-Critical Tasks with Recovery Policy
  2024/3/1
• Hybrid Trajectory Optimization for Autonomous Terrain Traversal of Articulated Tracked Robots
  2023/6/1
• Visuotactile Sensor Enabled Pneumatic Device Towards Compliant Oropharyngeal Swab Sampling
  2023/5/1