arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2610.21082cs.AI
收藏

SciΠvot:评估科学自动研究智能体关键节点的首席研究员判断能力的基准

Sci$Π$vot: Benchmarking Principal-Investigator Judgment at Pivotal Moments of Scientific AutoResearch Agents

University of Maryland, College Park(马里兰大学帕克分校);Gradient Network;Stanford University(斯坦福大学);The Chinese University of Hong Kong(香港中文大学);Dartmouth College(达特茅斯学院);University of California, Los Angeles(加州大学洛杉矶分校)
Youran Sun, Xingyu Ren, Chugang Yi, Kejia Zhang, Di Wu, Yingkang Cao, Allen Zhou, Jianping Ye, Suying Liu, Jiaxuan Guo, Jianda Du, Weicheng Ye, Kecheng Liu, Hef… 展开作者

Youran Sun, Xingyu Ren, Chugang Yi, Kejia Zhang, Di Wu, Yingkang Cao, Allen Zhou, Jianping Ye, Suying Liu, Jiaxuan Guo, Jianda Du, Weicheng Ye, Kecheng Liu, Hefan Zhang, Haixin Wang, Jie Xiao, Eric Yang, Haizhao Yang

AI总结:

本研究构建SciΠvot基准,评估LLM智能体的科学判断,发现前沿模型在该基准上表现低于随机水平,最高准确率63.5%,低于人类的73.4%,该基准可用于系统评估LLM智能体的科学判断能力。

摘要:

大型语言模型(LLM)智能体可在极少人工监督下维持科学实验运行数月,但可靠的自主研究还需要科学判断。评估科学判断颇具难度,因为研究决策具有开放性且依赖项目背景。我们构建了SciΠvot,这是一个基于人工修正的248个科学判断案例的基准。该基准利用了六个月的研究记录,期间资深研究人员在六个科学领域的26个项目上运行LLM智能体,并反复干预以修正智能体的研究判断。在每个案例中,智能体仅使用修正前可用的项目记录作为上下文,在原始LLM计划与人工修正的替代方案之间做出选择。我们比较了LLM与人类的表现,并研究了推理努力程度、项目记录访问权限以及重新考虑的影响。在中等推理努力且可访问项目记录的情况下,模型系统地偏好原始LLM计划而非人工修正的替代方案。五款前沿模型中有四款在所选案例上的表现低于随机水平。在项目条件下,所有测试模型和推理努力设置中的最高准确率为63.5%,而新研究人员的准确率为73.4%。在中等努力程度下,所有模型在可访问项目记录时的准确率均低于不可访问时,且重新考虑并未持续提高整体准确率。即使有人工修正的替代方案可用,前沿模型仍难以应对所选研究决策,而SciΠvot支持对LLM智能体的科学判断进行系统评估。

英文摘要:

Large language model (LLM) agents can keep scientific experiments running for months with little human supervision, but reliable autonomous research also requires scientific judgment. Evaluating scientific judgment is difficult because research decisions are open-ended and depend on project context. We construct Sci$Π$vot, a 248-case benchmark of scientific judgment based on human corrections. The benchmark draws on records from six months of research in which senior researchers ran LLM agents on 26 projects across six scientific fields and repeatedly intervened to correct the agents' research judgments. In each case, an agent chooses between the original LLM plan and the human-corrected alternative, using only the project record available before the correction as context. We compare LLM and human performance and examine the effects of reasoning effort, access to the project record, and reconsideration. At medium reasoning effort with the project record, models systematically prefer the original LLM plan over the human-corrected alternative. Four of five frontier models perform below chance on the selected cases. The highest accuracy in the project condition across the tested models and reasoning effort settings is 63.5%, compared with 73.4% for fresh researchers. At medium effort, every model achieves lower accuracy with the project record than without it, and reconsideration does not consistently improve overall accuracy. Even with a human-corrected alternative available, frontier models struggle with the selected research decisions, while Sci$Π$vot supports systematic evaluation of scientific judgment for LLM agents.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑