Reproduce “Software Engineering for AI-Driven Science” on OpenResearch
Turn your favorite coding agents into research agents that review literature, develop hypotheses, and run experiments.
Universal app for Apple silicon and Intel. Requires macOS 11 or later.
About this paper
Lukas Twist, Xinjian Wu, Jingzhi Gong, Shuyin Ouyang, Ying Xiao, Benedetta Donato, Yaoqi Guo, Zhenpeng Chen and 3 more
The authors propose Software Engineering for AI-Driven Science, adapting software-engineering assurance across the path from research question to scientific claim. Its Scientific Assurance Chain links intent, experimental design, implementation, execution, evidence, and claim, preserving assumptions and provenance across stages. Six illustrative failures, including a plausible accuracy despite a different evaluation procedure, motivate the framework, which is proposed, not evaluated.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.