Research map / Vision-language and multimodal / Robot action and navigation
RoboPearls: Editable Video Simulation for Robot Manipulation
Tang Tao, Likui Zhang, Youpeng Wen, Kaidong Zhang, Jia-Wang Bian, Xia Zhou, Tianyi Yan, Kun Zhan, Peng Jia, Hefeng Wu, Liang Lin, Xiaodan Liang
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 |
|---|---|---|---|
| replace 2D IP2P editing with RoboPearls 3D simulation (ablation) | RVT · RLBench Stack Cups | success rate: 18.4 -> 37.7 | Improved |
| replace 2D IP2P editing with RoboPearls 3D simulation (ablation) | RVT · RLBench Put in Cupboard | success rate: 44.8 -> 55.5 | Improved |
| add VLM failure analysis to RoboPearls (ablation) | RoboPearls · RLBench Stack Cups | success rate: 24.7 -> 37.7 | Improved |
| add VLM failure analysis to RoboPearls (ablation) | RoboPearls · RLBench Put in Cupboard | success rate: 45.0 -> 55.5 | Improved |
| add VLM failure analysis to RoboPearls (ablation) | RoboPearls · RLBench Insert Peg | success rate: 16.5 -> 17.1 | No clear effect |
| add IP2P edited data to RVT (ablation) | RVT · RLBench Insert Peg | success rate: 11.0 -> 10.7 | No clear effect |
| train with RoboPearls simulation data | RVT · COLOSSEUM average | success rate: 51.7 -> 69.2 | Improved |
| train with RoboPearls simulation data | RVT-2 · COLOSSEUM average | success rate: 64.6 -> 75.4 | Improved |
| train with RoboPearls simulation data | RVT-2 · COLOSSEUM all perturbations | success rate: 36.1 -> 54.7 | Improved |
| train with RoboPearls simulation data | RVT · RLBench average | success rate: 62.4 -> 68.0 | Improved |
| train with RoboPearls simulation data | SAM2Act · RLBench average | success rate: 83.8 -> 88.5 | Improved |
| train RDT with RoboPearls simulation data | RDT · real robot Place in (unseen) | successes out of 20: 1 / 20 -> 12 / 20 | Improved |
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.
- SAM2Act — SAM2Act: Integrating Visual Foundation Model with A Memory Architecture for Robotic Manipulation (ICML 2025) · better on 3 of 3 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
- Zero-Shot Robotic Manipulation with Pre-Trained Image-Editing Diffusion Models (ICLR 2024 · similarity 0.89)
- Robots Pre-train Robots: Manipulation-Centric Robotic Representation from Large-Scale Robot Datasets (ICLR 2025 · similarity 0.89)
- RoboDreamer: Learning Compositional World Models for Robot Imagination (ICML 2024 · similarity 0.88)
- Predictive Inverse Dynamics Models are Scalable Learners for Robotic Manipulation (ICLR 2025 · similarity 0.87)
- Generalizable Domain Adaptation for Sim-and-Real Policy Co-Training (NeurIPS 2025 · similarity 0.87)
- UniDomain: Pretraining a Unified PDDL Domain from Real-World Demonstrations for Generalizable Robot Task Planning (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.
