Research map / Reinforcement learning / Visual and representation RL
RB2: Robotic Manipulation Benchmarking with a Twist
Sudeep Dasari, Jianren Wang, Joyce Hong, Shikhar Bahl, Yixin Lin, Austin S Wang, Abitha Thankaraj, Karanbir Singh Chahal, Berk Calli, Saurabh Gupta, David Held, Lerrel Pinto and 3 more
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 open-loop behavior cloning (BC-Open) instead of more complex baselines | RB2 manipulation tasks · Pouring, scooping, insertion, and zipping | rank | Improved |
| Use closed-loop behavior cloning instead of open-loop behavior cloning | RB2 · Pouring, scooping, insertion, and zipping | rank | Hurt |
| Add an LSTM to closed-loop behavior cloning | BC-Closed · Pouring, scooping, insertion, and zipping | rank | Hurt |
| Use MOReL offline reinforcement learning instead of behavior cloning | RB2 · All four tasks | task performance | Hurt |
| Add noise-augmented and random-interaction data to MOReL's training dataset (ablation) | MOReL · Pouring, Franka Lab 1 | task performance | No clear effect |
Closest papers by meaning
- SAM2Act: Integrating Visual Foundation Model with A Memory Architecture for Robotic Manipulation (ICML 2025 · similarity 0.86)
- DexFlyWheel: A Scalable and Self-improving Data Generation Framework for Dexterous Manipulation (NeurIPS 2025 · similarity 0.86)
- RoboProcessBench: Benchmarking Process-Aware Understanding in Visual-Language Robotic Manipulation (NeurIPS 2026 · similarity 0.86)
- ManipTrans: Efficient Dexterous Bimanual Manipulation Transfer via Residual Learning (CVPR 2025 · similarity 0.86)
- SCaR: Refining Skill Chaining for Long-Horizon Robotic Manipulation via Dual Regularization (NeurIPS 2024 · similarity 0.86)
- Grasp-Then-Plan with Failure Attribution: A Closed Two-Stage Framework for Precise and Generalizable Robotic Manipulation (NeurIPS 2026 · similarity 0.86)
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.
