arXiv：cs.LG· Xuchen Gong, Shane Gu, Haokun Liu, Dixi Yao, Chenhao Tan, Tian Li·· 4 小时前AI 评分34
ResearchTrails：从人类研究决策轨迹中学习科学探索
Learning Scientific Exploration from Human Research Decision Trajectories
AI 导读
研究者提出 ResearchTrails，一个从 Git 仓库 commit 历史构建的人类研究轨迹数据集，commit 记录被用作研究探索过程的代理，可捕捉方法、实验与消融的连续变化。该数据集包含论文未披露的中间研究决策信号，并已验证两类用途：测试时检索人类研究经验作为外部技能，以及在研究轨迹上训练模型以提升对新研究决策的泛化能力。
正文
Abstract:A key challenge in building AI systems for scientific research is enabling $\textit{scientific exploration}$: the systematic process of investigating unknown phenomena or ideas to gain new knowledge through sequences of research decisions and actions. Yet this process is largely missing from existing scientific corpora; for example, research papers primarily record final outcomes rather than the trajectories that produced them. In this work, we introduce $\textbf{ResearchTrails}$, a dataset of $\textbf{human research trajectories constructed from Git repositories}$, where $\textbf{commit histories}$ serve as proxies for research exploration. We develop an automated and scalable pipeline that extracts structured research trajectories from repository commits, capturing successive changes to methods, experiments, and ablations. We characterize the resulting dataset and show that these trajectories contain meaningful signals about intermediate research decisions beyond what final papers reveal. We further demonstrate utilities of ResearchTrails in multiple use cases, including retrieving human research experience as external skills at test time and training models on research trajectories to improve generalization to new research decisions. Our results suggest a path toward AI systems that learn not only from the products of science, but from the evolving process of discovery itself.
| Subjects: | Machine Learning (cs.LG); Computation and Language (cs.CL) |
| Cite as: | arXiv:2610.07184 [cs.LG] |
| (or arXiv:2610.07184v1 [cs.LG] for this version) | |
| https://doi.org/10.48550/arXiv.2610.07184 arXiv-issued DOI via DataCite (pending registration) |
Submission history
From: Xuchen Gong [view email]
[v1]
Mon, 5 Oct 2026 18:05:12 UTC (843 KB)
来源：arXiv：cs.LG · arxiv.org