Saved in:
| Main Authors: | , , , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | https://arxiv.org/abs/2502.16747 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866909617324294144 |
|---|---|
| author | Nguyen, Dai Quoc Hoang, Cong Duy Vu Vu, Duy Tangari, Gioacchino Vu, Thanh Tien Dharmasiri, Don Li, Yuan-Fang Duong, Long |
| author_facet | Nguyen, Dai Quoc Hoang, Cong Duy Vu Vu, Duy Tangari, Gioacchino Vu, Thanh Tien Dharmasiri, Don Li, Yuan-Fang Duong, Long |
| contents | Open-weight large language models (LLMs) have significantly advanced performance in the Natural Language to SQL (NL2SQL) task. However, their effectiveness diminishes when dealing with large database schemas, as the context length increases. To address this limitation, we present SQLong, a novel and efficient data augmentation framework designed to enhance LLM performance in long-context scenarios for the NL2SQL task. SQLong generates augmented datasets by extending existing database schemas with additional synthetic CREATE TABLE commands and corresponding data rows, sampled from diverse schemas in the training data. This approach effectively simulates long-context scenarios during finetuning and evaluation. Through experiments on the Spider and BIRD datasets, we demonstrate that LLMs finetuned with SQLong-augmented data significantly outperform those trained on standard datasets. These imply SQLong's practical implementation and its impact on improving NL2SQL capabilities in real-world settings with complex database schemas. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2502_16747 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | SQLong: Enhanced NL2SQL for Longer Contexts with LLMs Nguyen, Dai Quoc Hoang, Cong Duy Vu Vu, Duy Tangari, Gioacchino Vu, Thanh Tien Dharmasiri, Don Li, Yuan-Fang Duong, Long Computation and Language Artificial Intelligence Machine Learning Software Engineering Open-weight large language models (LLMs) have significantly advanced performance in the Natural Language to SQL (NL2SQL) task. However, their effectiveness diminishes when dealing with large database schemas, as the context length increases. To address this limitation, we present SQLong, a novel and efficient data augmentation framework designed to enhance LLM performance in long-context scenarios for the NL2SQL task. SQLong generates augmented datasets by extending existing database schemas with additional synthetic CREATE TABLE commands and corresponding data rows, sampled from diverse schemas in the training data. This approach effectively simulates long-context scenarios during finetuning and evaluation. Through experiments on the Spider and BIRD datasets, we demonstrate that LLMs finetuned with SQLong-augmented data significantly outperform those trained on standard datasets. These imply SQLong's practical implementation and its impact on improving NL2SQL capabilities in real-world settings with complex database schemas. |
| title | SQLong: Enhanced NL2SQL for Longer Contexts with LLMs |
| topic | Computation and Language Artificial Intelligence Machine Learning Software Engineering |
| url | https://arxiv.org/abs/2502.16747 |