SIRIUS-SQL: Anchoring Multi-Candidate Text-to-SQL in Execution Feedback
Fuente:
arXiv
Salvato in:
| Autori principali: | , , , , , , , , , , |
|---|---|
| Natura: | Preprint |
| Pubblicazione: |
2026
|
| Soggetti: | |
| Accesso online: | |
| Tags: |
Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
|
| _version_ | 1866911738311475200 |
|---|---|
| author | Luo, Leo Xie, Haining Shen, Siqi Ma, Zhipeng Ling, Rui Xu, Hang Jiang, Hefeng Chen, Dingwei Li, Yang Chen, Peng Jiang, Jie |
| author_facet | Luo, Leo Xie, Haining Shen, Siqi Ma, Zhipeng Ling, Rui Xu, Hang Jiang, Hefeng Chen, Dingwei Li, Yang Chen, Peng Jiang, Jie |
| contents | Text-to-SQL on complex schemas is unreliable on a single pass, so recent systems generate multiple SQL candidates and let voting filter out errors. Yet voting alone is not enough, because the multi-candidate recipe has three coupled weaknesses: 1) sampling more from a single generator produces increasingly redundant candidates, 2) existing pipelines apply one generic correction to every non-clean execution result, while runtime errors, timeouts, and empty results each indicate a different distance from correctness, and 3) existing selectors rely on a single angle such as result-majority voting or pairwise SQL comparison, missing what other angles would have caught. We present SIRIUS-SQL, which addresses all three weaknesses. A difficulty-smoothing RL recipe trains SIRIUS-32B to generate diverse executable SQL candidates, paired with a generalist LLM that fills in gaps left by the specialist. An execution-grounded lifecycle classifies each outcome and applies targeted repair before candidates re-enter the pool. A confidence-gated hybrid selector combines execution-result agreement with pairwise SQL-form judgment, escalating only near-tied cases to a deterministic structural check. SIRIUS-SQL reaches 75.88% on BIRD dev and 91.20% on SPIDER test. Two of three generalist pairings surpass Agentar-Scale-SQL, the strongest published multi-candidate system on BIRD dev. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2606_01246 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | SIRIUS-SQL: Anchoring Multi-Candidate Text-to-SQL in Execution Feedback Luo, Leo Xie, Haining Shen, Siqi Ma, Zhipeng Ling, Rui Xu, Hang Jiang, Hefeng Chen, Dingwei Li, Yang Chen, Peng Jiang, Jie Artificial Intelligence Text-to-SQL on complex schemas is unreliable on a single pass, so recent systems generate multiple SQL candidates and let voting filter out errors. Yet voting alone is not enough, because the multi-candidate recipe has three coupled weaknesses: 1) sampling more from a single generator produces increasingly redundant candidates, 2) existing pipelines apply one generic correction to every non-clean execution result, while runtime errors, timeouts, and empty results each indicate a different distance from correctness, and 3) existing selectors rely on a single angle such as result-majority voting or pairwise SQL comparison, missing what other angles would have caught. We present SIRIUS-SQL, which addresses all three weaknesses. A difficulty-smoothing RL recipe trains SIRIUS-32B to generate diverse executable SQL candidates, paired with a generalist LLM that fills in gaps left by the specialist. An execution-grounded lifecycle classifies each outcome and applies targeted repair before candidates re-enter the pool. A confidence-gated hybrid selector combines execution-result agreement with pairwise SQL-form judgment, escalating only near-tied cases to a deterministic structural check. SIRIUS-SQL reaches 75.88% on BIRD dev and 91.20% on SPIDER test. Two of three generalist pairings surpass Agentar-Scale-SQL, the strongest published multi-candidate system on BIRD dev. |
| title | SIRIUS-SQL: Anchoring Multi-Candidate Text-to-SQL in Execution Feedback |
| topic | Artificial Intelligence |
| url | https://arxiv.org/abs/2606.01246 |