Reproduce “Probing Ambiguity: Adversarial Intent Resolution for Robust Text-to-SQL” 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
Youbing Yin, Xujun Peng, Muhammad S. Uddin, Erin Babinsky, Daben Liu, Anoop Kumar
AIR addresses ambiguous natural-language database questions by finding alternative, text-supported readings, translating them into SQL, and checking whether their results differ. One agent proposes an initial reading, while another searches for a defensible alternative; AIR asks a clarification only when executions diverge. On AMBROSIA, Claude AIR scores F1 0.679 versus IntentRL’s 0.618 with schema-only context, but IntentRL scores higher with full-database context, and AIR still requires database execution access.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.