titleBenchmarking AI Agents for Addressing Scientific Challenges Across Scales
authorsTianyu Liu, Allen Xin Wang, Antonia Panescu, Lisa Xinyi Chen, Wenxin Long, Xinyu Wei, Yueqian Jing, Ziyao Zeng, Jihang Chen, Sihan Jiang, Ziqing Wang, Siyi Gu, Siyu Chen, Xinyang Hu, Haoran Shao, Leqi Xu, Wangjie Zheng, Zhiyuan Cao, Ada Fang, Botao Yu, Kunyang Sun, Rex Ying, Arman Cohan, Qingyu Chen, Lingzhou Xue, Kaize Ding, Yuanqi Du, Wengong Jin, Zhuoran Yang, Marinka Zitnik, James Zou, Hua Xu, Hongyu Zhao
date2026-06-10
venuearXiv (preprint, not peer reviewed)
idarXiv:2606.12736
upvotes5 (Hugging Face Daily Papers, 2026-11-01)
linksabs pdf hf code
abstractAI agents are increasingly being developed to accelerate scientific discovery, yet their practical capabilities in real research settings remain poorly understood. Existing benchmarks for AI agents rarely capture the complexity, heterogeneity, and extended reasoning required by scientific work, whereas benchmarks for scientific tasks often reduce research to static, direct problems and provide limited support for interactive evaluation. Here, we introduce SciAgentArena, a systematic benchmark for evaluating AI agents in real-world scientific research scenarios drawn from emerging needs across multiple domains. SciAgentArena comprises approximately 200 tasks with stepwise verification and an interactive, agent-agnostic environment for assessing diverse AI agents. Using this benchmark, we find that current agents can contribute effectively to well-specified data-analysis workflows, particularly when the task structure and evaluation criteria are clear. However, their performance remains uneven across scientific contexts: agents struggle to generate genuinely novel insights, sustain self-directed exploration, and formulate robust solutions for open-ended research questions. We further characterize common failure modes across agents and identify opportunities for improving their reliability, autonomy, and scientific reasoning. Together, SciAgentArena provides a practical framework for measuring progress in AI agents for science and for guiding the design of future agents capable of addressing complex scientific challenges. Full codes, tasks, and datasets can be accessed via this link: https://sciagentarena.github.io/.

cite

@misc{liu2026benchmarking,
  title={Benchmarking AI Agents for Addressing Scientific Challenges Across Scales},
  author={Tianyu Liu and Allen Xin Wang and Antonia Panescu and Lisa Xinyi Chen and Wenxin Long and Xinyu Wei and Yueqian Jing and Ziyao Zeng and Jihang Chen and Sihan Jiang and Ziqing Wang and Siyi Gu and Siyu Chen and Xinyang Hu and Haoran Shao and Leqi Xu and Wangjie Zheng and Zhiyuan Cao and Ada Fang and Botao Yu and Kunyang Sun and Rex Ying and Arman Cohan and Qingyu Chen and Lingzhou Xue and Kaize Ding and Yuanqi Du and Wengong Jin and Zhuoran Yang and Marinka Zitnik and James Zou and Hua Xu and Hongyu Zhao},
  year={2026},
  eprint={2606.12736},
  archivePrefix={arXiv},
  url={https://arxiv.org/abs/2606.12736}
}