Reproduce “Data.Readme: Documenting Table Relationships for Diverse Data Tasks” 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
Jiaming Liang, Yeye He, Haoyu Dong, Zachary Ives, Surajit Chaudhuri
Data.Readme makes data lakes navigable by documenting table relationships alongside individual descriptions. It indexes confusable, joinable, and union-able tables, then lets agents retrieve entries and follow links during analysis. Across 96 data-discovery and data-analysis settings, it ranked highest in 92, including one discovery tie. These benchmarks use two benchmark-derived lakes, one augmented with synthetic tables.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.