Research map / Reinforcement learning / Visual and representation RL
Learning Score-based Grasping Primitive for Human-assisting Dexterous Grasping
Tianhao Wu, Mingdong Wu, Jiyao Zhang, Yunchong Gan, Hao Dong
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 |
|---|---|---|---|
| remove the RL residual policy (primitive only) (ablation) | Ours · seen category unseen instances | Success | Hurt |
| remove the score-based primitive guidance (GF) (ablation) | Ours · seen category unseen instances | Success | Hurt |
| remove the ar action module (ablation) | Ours · training | Success | Hurt |
| remove the as action module (ablation) | Ours · training | Success | Hurt |
| add ar module to ap (.) as (ablation) | primitive action · action module combinations | Success rate: 5.08 % -> 56.50 % | Improved |
| add as module to primitive action (ablation) | primitive action · action module combinations | Success rate: 19.49 % -> 5.08 % | Hurt |
| disable collisions for primitive action (ablation) | primitive action · action module combinations | Success rate: 19.49 % -> 55.60 % | Improved |
| remove intrinsic reward r_sim (ablation) | Ours · training | Success | Hurt |
| increase wrist observation noise from 0 to 5 deg/5 cm | Ours · unseen category instances | Success rate: 56.69% -> 44.35% | Hurt |
| increase wrist observation noise from 0 to 2 deg/2 cm | Ours · unseen category instances | Success rate: 56.69% -> 55.24% | No clear effect |
| Ours instead of ILAD | dexterous grasping policy · seen category unseen instances | Success (%): 45.9 -> 56.5 | Improved |
| Ours instead of ILAD | dexterous grasping policy · seen category | Translation (cm): 2.443 -> 2.131 | Improved |
Closest papers by meaning
- Toward Human-Like Grasp: Dexterous Grasping via Semantic Representation of Object-Hand (ICCV 2021 · similarity 0.89)
- ManipTrans: Efficient Dexterous Bimanual Manipulation Transfer via Residual Learning (CVPR 2025 · similarity 0.88)
- Grasp as You Say: Language-guided Dexterous Grasp Generation (NeurIPS 2024 · similarity 0.88)
- Cross-Embodiment Dexterous Grasping with Reinforcement Learning (ICLR 2025 · similarity 0.88)
- DexFuncGrasp: A Robotic Dexterous Functional Grasp Dataset Constructed from a Cost-Effective Real-Simulation Annotation System (AAAI 2024 · similarity 0.87)
- DexGrasp Anything: Towards Universal Robotic Dexterous Grasping with Physics Awareness (CVPR 2025 · similarity 0.87)
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.
