Reproduce “Sage-Taiyi: A Metacognitive Architecture for Scientific Discovery” 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
Weida Wang, Benteng Chen, Yigeng Jiang, Shuchen Pu, Weiping Shen, Zichao Liu, Yulin Su, Zitong Wang and 82 more
The authors present Sage-Taiyi, a research system that tracks evidence, assumptions, uncertainty, and scope to guide which targets and reasoning methods to pursue. In expert-guided cobalt-catalyzed carbonylation, condition optimization raised isolated yield from trace to 84%. On Taiyi-Bench, Sage-Taiyi scored 53.95 versus 47.85 for AI Scientist-v2; this full-system comparison does not isolate monitoring’s contribution.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.