titleOSWorld-Science: A Benchmark of Computer Use Agents for Learning and Using Scientific Software
authorsDingyuan Dai, Heli Qi, Lei Liu, Yinxi Li, Baiding Chen, Zijun Dou, Qingcheng Zeng, Qi Kang, Oliver Sun, Eric Wang, Bo Zhou, Haixin Wang, Yufan Du, Shi Bo, Ruihan Lin, Mengqi Yuan, Dunjie Lu, Steven Dillmann, Yiming Shi, Tina Su, Amy Xin, Minghao Liu, Xi Wang, Xu Huang, Ge Zhang, Pengyu Nie, Zhen Yang, Jie Tang, Juanzi Li, Weihao Xuan, Tianyu Liu
date2026-09-30
venuearXiv (preprint, not peer reviewed)
idarXiv:2609.39903
affiliationDiscoAILab
upvotes27 (Hugging Face Daily Papers, 2026-11-01)
linksabs pdf hf code
abstractScientific software presents a demanding test for computer-using agents based on visual language models (VLMs): completing a research workflow requires interpreting specialized interfaces, manipulating scientific objects, and producing verifiable results. We thus introduce OSWorld-Science, a benchmark and evaluation environment that combines scientifically meaningful tasks, artifact-based evaluation, and an efficient agent harness for studying computer use in the scientific domain. The benchmark contains 12 VLMs and 146 high-quality tasks across several scientific domains and software configurations, covering workflows such as molecular drawing and retrosynthesis, pathology image analysis, statistical computing, and physical simulation. Tasks are developed through expert proposals and iterative human--AI co-design, with selection guided by scientific value and difficulty. Task-specific execution-based evaluators inspect application states and generated artifacts, including molecular structures, segmentation masks, plots, and numerical results, and award partial credit for incomplete outcomes. Our special harness integrates model adapters, interaction-loop control, and trajectory logging to support comparisons of models and interaction strategies. Our results show that current state-of-the-art VLMs with a strong harness still face challenges in addressing key questions in the scientific domains. We also analyze the benchmarking results across multi-linguistics, reasoning efforts, context length and other factors and derive several important conclusions and directions to assist future development. Overall, we provide an integrated framework connecting expert-defined scientific goals to verifiable software outcomes, enabling systematic evaluation of both agent capabilities and harness design in scientific workflows.

cite

@misc{dai2026osworld,
  title={OSWorld-Science: A Benchmark of Computer Use Agents for Learning and Using Scientific Software},
  author={Dingyuan Dai and Heli Qi and Lei Liu and Yinxi Li and Baiding Chen and Zijun Dou and Qingcheng Zeng and Qi Kang and Oliver Sun and Eric Wang and Bo Zhou and Haixin Wang and Yufan Du and Shi Bo and Ruihan Lin and Mengqi Yuan and Dunjie Lu and Steven Dillmann and Yiming Shi and Tina Su and Amy Xin and Minghao Liu and Xi Wang and Xu Huang and Ge Zhang and Pengyu Nie and Zhen Yang and Jie Tang and Juanzi Li and Weihao Xuan and Tianyu Liu},
  year={2026},
  eprint={2609.39903},
  archivePrefix={arXiv},
  url={https://arxiv.org/abs/2609.39903}
}