[论文] ScienceIDE: Turning World's Scientific Codebase into Agent Learnable E...

论文概要

研究领域: NLP 作者: Hejia Geng, Zesen Huang, Haoyang Li, Wenbin Li, Koutian Wu, Zihan Zhou, Yuanbo Pang, Weihao Liu, Zigong Xu, Zhiping Li, Zongzheng Zhang, Chuanfei Dong, Jiankai Sun, Tianzhe Zheng, Fengyu Xie, Yue Ma, Yueheng Shi, Tong Xie, Zonglin Di, Xianrong Liu, Qucheng Gao, Yimin Liu, Jiaming Pan, Sheng Huang, Xiao-Han Ma, Lanqing Yuan, Zhenlin Zhu, Ziang Liu, Ziyang Xu, Junkai Wang, Kangkai Liang, Jiayi Xian, Zehong Zhao, Liuwei Xu, Jingxu Xie, Peijin Zhang, Qiang Gao, Chengyi Xing, Zhe Zhao, Xi Wang, Yaopeng Xing, Xing Meng, Zhenfei Yin, Yingcheng Wu, Ling Yang 发布时间: 2026-09-16 arXiv: 2609.19134

中文摘要

科学代码库将数十年的人类知识编码为可执行的模型、方法与工具。然而碎片化的工具链、隐含的领域惯例与专门的正确性标准，使这些知识难以转化为可靠的学习经验——我们将这一挑战称为"科学经验瓶颈"。我们引入 ScienceIDE：将全球科学代码转化为科学智能体可编程环境的基础设施。在专家定义的科学案例与验收标准指导下，智能体将代码仓库转化为可执行环境，支持任务生成、执行与科学验证。这些环境为监督微调、强化学习和评估提供共同基础。利用验证过的交互轨迹，我们训练了 PhAI-IDE-72B、PhAI-IDE-9B 与 PhAI-IDE-4B。该模型家族在 held-out 科学代码修复以及代码、推理、知识等通用基准上均有提升，为科学经验向更广泛能力的正迁移提供了证据。ScienceIDE 为智能体学习与科学实践的集成工作空间奠定基础，使人类的科学软件成为发展科学智能的共同基底。代码已开源。

原文摘要

Scientific code repositories encode decades of human knowledge in executable models, methods, and tools. Yet fragmented toolchains, implicit domain conventions, and specialized correctness criteria make this knowledge difficult to convert into reliable learning experience-a challenge we call the scientific experience bottleneck. We introduce ScienceIDE, infrastructure for turning the world's scientific code into programmable environments for scientific agents. Guided by expert-defined scientific cases and acceptance criteria, agents transform repositories into executable environments that support task generation, execution, and scientific verification. These environments provide a shared foundation for supervised fine-tuning, reinforcement learning, and evaluation. Using verified interacti...

#论文 #arXiv #NLP #小凯