You Only Read Once (YORO): Learning to Internalize Database Knowledge for Text-to-SQL
Fuente:
arXiv
Salvato in:
| Autori principali: | , , , , , , , |
|---|---|
| Natura: | Preprint |
| Pubblicazione: |
2024
|
| Soggetti: | |
| Accesso online: | |
| Tags: |
Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
|
| _version_ | 1866917779413663744 |
|---|---|
| author | Kobayashi, Hideo Lan, Wuwei Shi, Peng Chang, Shuaichen Guo, Jiang Zhu, Henghui Wang, Zhiguo Ng, Patrick |
| author_facet | Kobayashi, Hideo Lan, Wuwei Shi, Peng Chang, Shuaichen Guo, Jiang Zhu, Henghui Wang, Zhiguo Ng, Patrick |
| contents | While significant progress has been made on the text-to-SQL task, recent solutions repeatedly encode the same database schema for every question, resulting in unnecessary high inference cost and often overlooking crucial database knowledge. To address these issues, we propose You Only Read Once (YORO), a novel paradigm that directly internalizes database knowledge into the parametric knowledge of a text-to-SQL model during training and eliminates the need for schema encoding during inference. YORO significantly reduces the input token length by 66%-98%. Despite its shorter inputs, our empirical results demonstrate YORO's competitive performances with traditional systems on three benchmarks as well as its significant outperformance on large databases. Furthermore, YORO excels in handling questions with challenging value retrievals such as abbreviation. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2409_12172 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | You Only Read Once (YORO): Learning to Internalize Database Knowledge for Text-to-SQL Kobayashi, Hideo Lan, Wuwei Shi, Peng Chang, Shuaichen Guo, Jiang Zhu, Henghui Wang, Zhiguo Ng, Patrick Computation and Language While significant progress has been made on the text-to-SQL task, recent solutions repeatedly encode the same database schema for every question, resulting in unnecessary high inference cost and often overlooking crucial database knowledge. To address these issues, we propose You Only Read Once (YORO), a novel paradigm that directly internalizes database knowledge into the parametric knowledge of a text-to-SQL model during training and eliminates the need for schema encoding during inference. YORO significantly reduces the input token length by 66%-98%. Despite its shorter inputs, our empirical results demonstrate YORO's competitive performances with traditional systems on three benchmarks as well as its significant outperformance on large databases. Furthermore, YORO excels in handling questions with challenging value retrievals such as abbreviation. |
| title | You Only Read Once (YORO): Learning to Internalize Database Knowledge for Text-to-SQL |
| topic | Computation and Language |
| url | https://arxiv.org/abs/2409.12172 |