Research map / Reinforcement learning / Visual and representation RL
UniBYD: A Unified Framework for Learning Robotic Manipulation Across Embodiments Beyond Imitation of Human Demonstrations
Tingyu Yuan, Biaoliang Guan, Wen Ye, Ziyan Tian, Yi Yang, Weijie Zhou, Zhaowen Li, Yan Huang, Peng Wang, Chaoyang Zhao, Jinqiao Wang
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| add shadow engine (ablation) | base imitation RL · UniManip 5-finger (1 hand) | success rate (%): 13.56 -> 55.11 | Improved |
| add shadow engine (ablation) | base imitation RL · UniManip 2-finger (1 hand) | success rate (%): 24.94 -> 67.06 | Improved |
| add shadow engine (ablation) | base imitation RL · UniManip 3-finger (1 hand) | success rate (%): 19.56 -> 51.06 | Improved |
| add shadow engine (ablation) | base imitation RL · UniManip 5-finger (2 hands) | success rate (%): 33.46 -> 43.33 | Improved |
| add goal reward (ablation) | base imitation RL · UniManip 5-finger (1 hand) | success rate (%): 13.56 -> 63.17 | Improved |
| add goal reward (ablation) | base imitation RL · UniManip 5-finger (2 hands) | success rate (%): 33.46 -> 47.01 | Improved |
| add goal reward (ablation) | base imitation RL · UniManip 5-finger (2 hands) | adaptation score: 3.20 -> 6.27 | Improved |
| add loss synergy and counterbalancing (entropy + bound loss) on top of goal reward (ablation) | +GR · UniManip 5-finger (1 hand) | success rate (%): 63.17 -> 65.39 | Improved |
| add loss synergy and counterbalancing on top of goal reward (ablation) | +GR · UniManip 2-finger (1 hand) | success rate (%): 56.19 -> 61.50 | Improved |
| add shadow engine to goal-reward and LSC variant (full UniBYD) (ablation) | +GR+LSC · UniManip 5-finger (1 hand) | success rate (%): 65.39 -> 85.67 | Improved |
| add shadow engine to goal-reward and LSC variant (full UniBYD) (ablation) | +GR+LSC · UniManip 5-finger (2 hands) | success rate (%): 54.17 -> 57.67 | Improved |
| add shadow engine to goal-reward and LSC variant (full UniBYD) (ablation) | +GR+LSC · UniManip 3-finger (1 hand) | position error (cm): 0.87 -> 0.89 | Hurt |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.
- ManipTrans — ManipTrans: Efficient Dexterous Bimanual Manipulation Transfer via Residual Learning (CVPR 2025) · better on 4 of 4 shared results
Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.
Closest papers by meaning
- Cross-Embodiment Dexterous Grasping with Reinforcement Learning (ICLR 2025 · similarity 0.90)
- ManipTrans: Efficient Dexterous Bimanual Manipulation Transfer via Residual Learning (CVPR 2025 · similarity 0.90)
- Universal Actions for Enhanced Embodied Foundation Models (CVPR 2025 · similarity 0.88)
- HandEdit: A Unified Benchmark for Egocentric Human-to-Robot Dexterous Hand Image Editing (NeurIPS 2026 · similarity 0.88)
- UniGraspTransformer: Simplified Policy Distillation for Scalable Dexterous Robotic Grasping (CVPR 2025 · similarity 0.88)
- ManiSkill-HAB: A Benchmark for Low-Level Manipulation in Home Rearrangement Tasks (ICLR 2025 · similarity 0.88)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
