Research map / 3D vision / Human pose estimation
ManiSkill: Generalizable Manipulation Skill Benchmark with Large-Scale Demonstrations
Tongzhou Mu, Zhan Ling, Fanbo Xiang, Derek Cathera Yang, Xuanlin Li, Stone Tao, Zhiao Huang, Zhiwei Jia, Hao Su
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 |
|---|---|---|---|
| Use PointNet + Transformer instead of PointNet with behavior cloning (ablation) | BC policy · OpenCabinetDrawer single environment | average success rate: 0.68 -> 0.85 | Improved |
| Use PointNet + Transformer instead of PointNet with behavior cloning (ablation) | BC policy · OpenCabinetDrawer single environment | average success rate: 0.76 -> 0.90 | Improved |
| Increase demonstration trajectories from 10 to 1000 (ablation) | PointNet + Transformer, BC · OpenCabinetDrawer single environment | average success rate: 0.16 -> 0.90 | Improved |
| Use BC instead of offline RL (ablation) | PointNet + Transformer policy · OpenCabinetDrawer single environment, 300 demos | average success rate: 0.45 -> 0.85; 0.31 -> 0.85 | Improved |
| Use PointNet + Transformer instead of PointNet (ablation) | BC policy · MoveBucket test | mean success rate: 0.02±0.01 -> 0.08±0.01 | Improved |
| Use PointNet + Transformer instead of PointNet (ablation) | BC policy · PushChair test | mean success rate: 0.09±0.02 -> 0.08±0.01 | Hurt |
| Train SAC on more cabinet objects rather than one cabinet (ablation) | SAC demonstration-generation agent · OpenCabinetDrawer | success rate: 100% -> 82% -> 2% -> 0% | Hurt |
| Add positional encoding to the Transformer (ablation) | PointNet + Transformer · ManiSkill baseline policies | policy performance | Hurt |
Closest papers by meaning
- DynaRend: Learning 3D Dynamics via Masked Future Rendering for Robotic Manipulation (NeurIPS 2025 · similarity 0.88)
- GENMANIP: LLM-driven Simulation for Generalizable Instruction-Following Manipulation (CVPR 2025 · similarity 0.88)
- Robots Pre-train Robots: Manipulation-Centric Robotic Representation from Large-Scale Robot Datasets (ICLR 2025 · similarity 0.88)
- RobotSmith: Generative Robotic Tool Design for Acquisition of Complex Manipulation Skills (NeurIPS 2025 · similarity 0.88)
- AffordDP: Generalizable Diffusion Policy with Transferable Affordance (CVPR 2025 · similarity 0.88)
- Exploring the Limits of Vision-Language-Action Manipulation in Cross-task Generalization (NeurIPS 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.
