一句话：提出规则驱动并结合查询归约的 SQL 方言翻译方法，将复杂查询拆解以缓解 LLM 处理长复杂 SQL 的困难。
将基于 RDBMS 的应用迁移到云端时，跨数据库管理系统的 SQL 方言翻译至关重要。传统翻译工具依赖人工编写的规则，每支持一种新的 RDBMS 或方言都需要大量人工投入；大语言模型虽可辅助方言翻译，却在冗长、复杂的 SQL 查询上表现不佳。RISE 采用规则驱动的方式，并通过查询归约（query reduction）把长而复杂的 SQL 查询拆解为更小的部分，使规则与模型各自处理其擅长的环节，从而在保持翻译过程可控的同时降低对 LLM 处理复杂整体的依赖。所给摘要未报告具体的实验设置与量化结果。
Translating SQL dialects across different relational database management systems (RDBMSs) is crucial for migrating RDBMS-based applications to the cloud. Traditional SQL dialect translation tools rely on manually-crafted rules, necessitating significant manual effort to support new RDBMSs and dialects. Although large language models (LLMs) can assist in translating SQL dialects, they often struggle with lengthy and complex SQL queries.
@inproceedings{paperbot11555,
title = {RISE: Rule-Driven SQL Dialect Translation via Query Reduction},
author = {Xudong Xie and Yuwei Zhang and Wensheng Dou and Yu Gao and Ziyu Cui and Jiansen Song and Rui Yang and Jun Wei},
booktitle = {ICSE},
year = {2026},
doi = {10.1145/3744916.3773257}
}