Reproduce “ARCS: Towards Precise Text-to-SQL via Structured Disambiguation” 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
Yihao Hu, Yanlin Feng, Naoki Otani, Nikita Bhutani
ARCS helps users resolve ambiguous text-to-SQL questions by presenting separate choices for finite interpretations and adjustable values and operators for thresholds, then generating or updating SQL from their selections. On the tested comparison, structured disambiguation achieved 42.44% execution accuracy with 18 simulator-effort units, versus 43.73% and 34 for conversational clarification; these units are not human effort or elapsed time.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.