SpotIt+: Verification-based Text-to-SQL Evaluation with Database Constraints
Fuente:
arXiv
Gespeichert in:
| Hauptverfasser: | , , , , , |
|---|---|
| Format: | Preprint |
| Veröffentlicht: |
2026
|
| Schlagworte: | |
| Online-Zugang: | |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| _version_ | 1866917481436676096 |
|---|---|
| author | Tremante, Andrew He, Yang Klopfenstein, Rocky Wang, Yuepeng Narodytska, Nina Wu, Haoze |
| author_facet | Tremante, Andrew He, Yang Klopfenstein, Rocky Wang, Yuepeng Narodytska, Nina Wu, Haoze |
| contents | We present SpotIt+, an open-source tool for evaluating Text-to-SQL systems via bounded equivalence verification. Given a generated SQL query and the ground truth, SpotIt+ actively searches for database instances that differentiate the two queries. To ensure that the generated counterexamples reflect practically relevant discrepancies, we introduce a best-effort constraint-mining pipeline that combines rule-based specification mining with LLM-based validation over example databases. Experimental results on the BIRD dataset show that the mined constraints enable SpotIt+ to generate more realistic differentiating databases, while preserving its ability to efficiently uncover numerous discrepancies between generated and gold SQL queries that are missed by standard test-based evaluation. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2603_04334 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | SpotIt+: Verification-based Text-to-SQL Evaluation with Database Constraints Tremante, Andrew He, Yang Klopfenstein, Rocky Wang, Yuepeng Narodytska, Nina Wu, Haoze Databases Artificial Intelligence Logic in Computer Science Programming Languages We present SpotIt+, an open-source tool for evaluating Text-to-SQL systems via bounded equivalence verification. Given a generated SQL query and the ground truth, SpotIt+ actively searches for database instances that differentiate the two queries. To ensure that the generated counterexamples reflect practically relevant discrepancies, we introduce a best-effort constraint-mining pipeline that combines rule-based specification mining with LLM-based validation over example databases. Experimental results on the BIRD dataset show that the mined constraints enable SpotIt+ to generate more realistic differentiating databases, while preserving its ability to efficiently uncover numerous discrepancies between generated and gold SQL queries that are missed by standard test-based evaluation. |
| title | SpotIt+: Verification-based Text-to-SQL Evaluation with Database Constraints |
| topic | Databases Artificial Intelligence Logic in Computer Science Programming Languages |
| url | https://arxiv.org/abs/2603.04334 |