CLARITY: A Framework and Benchmark for Conversational Language Ambiguity and Unanswerability in Interactive NL2SQL Systems

TL;DR

Clarity introduces a comprehensive benchmark for evaluating NL2SQL systems' ability to handle ambiguous and unanswerable queries in interactive scenarios, revealing significant performance challenges.

Contribution

It presents a novel framework for generating diverse, multi-faceted ambiguities in NL2SQL benchmarks and evaluates existing systems' robustness against these challenges.

Findings

Leading NL2SQL systems perform poorly under multi-faceted ambiguity.

Most systems detect ambiguity but struggle with localization and resolution.

Performance degradation is significant on benchmark datasets.

Abstract

NL2SQL systems deployed in industry settings often encounter ambiguous or unanswerable queries, particularly in interactive scenarios with incomplete user clarification. Existing benchmarks typically assume a single source of ambiguity and rely on user interaction for resolution, overlooking realistic failure modes. We introduce Clarity, a framework for automatically generating an NL2SQL benchmark with multi-faceted ambiguities and diverse user behaviors across both single- and multi-turn settings. Using a constraint-driven pipeline, Clarity transforms executable SQL into ambiguous queries, augmented with grounded conversational continuations and schema-level metadata. Empirical evaluation on Spider and BIRD shows that leading NL2SQL systems, including those based on strong LLMs, suffer significant performance degradation under multi-faceted ambiguity. While these systems often detect ambiguity, they struggle to accurately localize and resolve the underlying schema-level sources. Our results highlight the need for more robust ambiguity detection and resolution in industry-grade NL2SQL systems.