Reproduce “Learning a Neural Semantic Parser from User Feedback” 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
Srinivasan Iyer, Ioannis Konstas, Alvin Cheung, Jayant Krishnamurthy, Luke Zettlemoyer
The authors present a database interface that translates questions directly into SQL and uses deployment feedback to select examples for annotation. The model can add accepted predictions to training data, while crowd workers correct errors. In a three-stage academic-database study, user ratings of correct or incomplete rose from 25% to 53.7% to 63.5%; these ratings are not independently verified accuracy.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.