Research map / Vision-language and multimodal / Robot action and navigation
AutoBio: A Simulation and Benchmark for Robotic Automation in Digital Biology Laboratory
Zhiqian Lan, Yuxuan Jiang, Ruiqi Wang, Xuanbing Xie, Rongkui Zhang, Yicheng Zhu, LI PEIHANG, Tianshuo Yang, Tianxing Chen, Haoyu Gao, Xiaokang Yang, Xuelong Li 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 |
|---|---|---|---|
| increase demonstrations from 20 to 100 | pi0 · Aspirate with pipette | success rate (%): 4.3 -> 42.7 | Improved |
| increase demonstrations from 20 to 100 | pi0 · Transfer centrifuge tube | success rate (%): 2.0 -> 40.7 | Improved |
| increase demonstrations from 20 to 100 | pi0 · Unscrew centrifuge tube cap | success rate (%): 1.3 -> 21.3 | Improved |
| increase demonstrations from 20 to 100 | pi0 · Pick up centrifuge tube | success rate (%): 13.3 -> 53.7 | Improved |
| increase demonstrations from 20 to 100 | RDT · Aspirate with pipette | success rate (%): 0.0 -> 0.3 | No clear effect |
| increase demonstrations from 20 to 100 | RDT · Unscrew centrifuge tube cap | success rate (%): 2.0 -> 2.7 | No clear effect |
| increase demonstrations from 20 to 100 | RDT · Load centrifuge rotor | success rate (%): 1.7 -> 1.0 | No clear effect |
| increase demonstrations from 20 to 100 | RDT · Pick up centrifuge tube | success rate (%): 45.3 -> 57.7 | Mixed |
| use pi0 instead of RDT on hard task | pi0 vs RDT · Operate thermal mixer panel (100 demos) | success rate (%): 1.6 -> 7.5 | Improved |
| use RDT instead of pi0 on easy task with 20 demos | RDT vs pi0 · Open thermal cycler lid (20 demos) | success rate (%): 73.0 -> 100.0 | Improved |
| use pi0 instead of RDT for concatenated Open-Close | pi0 vs RDT · Open-Close concatenation | success rate (%): 8.7 -> 87.3 | Improved |
| concatenate Close-Open | pi0 and RDT · Close-Open concatenation | success rate (%): 3.0 and 4.3 | Hurt |
Closest papers by meaning
- VLA-REPLICA: A Low-Cost, Reproducible Benchmark for Real-World Evaluation of Vision-Language-Action Models (NeurIPS 2026 · similarity 0.88)
- LabUtopia: High-Fidelity Simulation and Hierarchical Benchmark for Scientific Embodied Agents (NeurIPS 2025 · similarity 0.87)
- Robotic Visual Instruction (CVPR 2025 · similarity 0.87)
- Exploring the Limits of Vision-Language-Action Manipulation in Cross-task Generalization (NeurIPS 2025 · similarity 0.87)
- RoboCerebra: A Large-scale Benchmark for Long-horizon Robotic Manipulation Evaluation (NeurIPS 2025 · similarity 0.87)
- CaP-X: A Framework for Benchmarking and Improving Coding Agents for Robot Manipulation (ICML 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.
