Reproduce “IdeaTaste: Do LLMs Share the Research Taste of Human Researchers?” 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
Shanshan Zhong, Chen Xu, Jingjie Ning, Zhengzhong Liu, Aviral Kumar, Chenyan Xiong
IdeaTaste tests whether AI research agents make choices resembling those of human communities. It generates proposals from literature frozen at the end of 2025, then compares their citations and ideas with papers published in January–August 2026. Agents reach about 26% of human level on citation matching and 18% on idea matching, but publication similarity measures community choices, not objective research quality.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.