Reproduce “Iteration Without Elaboration: A Simple ReAct Architecture Suffices for Text-to-SQL Generation” 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
Jian Lu, Haiwei Yu, Raymond M Xiong, Anru Zhang, Danyang Zhuo
ReAct-SQL translates natural-language database questions by probing live data and incrementally composing queries from 12 typed relational operations, with validation and execution feedback. On corrected BIRD mini-dev and EHR-SQL, it reaches 84.5% and 73.9% accuracy, respectively, using a fixed LLM judge’s MATCH decisions; these results cover 498 and 934 questions, not original BIRD exact-match scoring.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.