Kui Jia

The Chinese University of Hong Kong, Shenzhen

収録論文 21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

• RoboSynChallenge: 合成操作スキルの一般化による実世界の巧みな操作の習得
  操作/ベンチマーク2026/8/12

  実世界データ不足を補うため、大規模合成データ生成と標準化された実世界評価を組み合わせた操作ポリシーの一般化を競うベンチマークを提案した論文。

• Vid2WAM: ビデオ拡散事前知識を世界行動モデルへ蒸留する
  VLA2026/8/9

  大規模ビデオ基盤モデルの拡散事前知識をコンパクトな世界行動モデルに蒸留し、専門家デモの少ない環境でもロボットの汎化性能とデータ効率を向上させる手法を提案した。

• RoboSynChallenge: Mastering Real-World Dexterity via Generalizing Synthesized Manipulation Skills
  2026/8/1
• Vid2WAM: Distilling Video Diffusion Priors into World Action Models
  2026/8/1
• Dexterity-BEV: Aligning 3D World and Actions for Generalizable Robot Policies Learning
  2026/6/1
• From Reaction to Anticipation: Proactive Failure Recovery through Agentic Task Graph for Robotic Manipulation
  2026/5/1
• EVA: Aligning Video World Models with Executable Robot Actions via Inverse Dynamics Rewards
  2026/3/1
• PAct: Part-Decomposed Single-View Articulated Object Generation
  2026/2/1
• One-Shot Real-World Demonstration Synthesis for Scalable Bimanual Manipulation
  2025/12/1
• Embracing Evolution: A Call for Body-Control Co-Design in Embodied Humanoid Robot
  2025/10/1
• Toward Humanoid Brain-Body Co-design: Joint Optimization of Control and Morphology for Fall Recovery
  2025/10/1
• BiNoMaP: Learning Category-Level Bimanual Non-Prehensile Manipulation Primitives
  2025/9/1
• VLBiMan: Vision-Language Anchored One-Shot Demonstration Enables Generalizable Bimanual Robotic Manipulation
  2025/9/1
• ImaginationPolicy: Towards Generalizable, Precise and Reliable End-to-End Policy for Robotic Manipulation
  2025/9/1
• SignBot: Learning Human-to-Humanoid Sign Language Interaction
  2025/5/1
• Real-Time Verification of Embodied Reasoning for Generative Skill Acquisition
  2025/5/1
• Prof. Robot: Differentiable Robot Rendering Without Static and Self-Collisions
  2025/3/1
• GAT-Grasp: Gesture-Driven Affordance Transfer for Task-Aware Robotic Grasping
  2025/3/1
• HWC-Loco: A Hierarchical Whole-Body Control Approach to Robust Humanoid Locomotion
  2025/3/1
• You Only Teach Once: Learn One-Shot Bimanual Robotic Manipulation from Video Demonstrations
  2025/1/1
• W-PoseNet: Dense Correspondence Regularized Pixel Pair Pose Regression
  2019/12/1