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. In this paper, we propose RISE, a novel LLM-based SQL dialect translation approach that can accurately handle lengthy and complex SQL queries. Given a complex source query 𝑄𝑐 that contains a SQL dialect 𝑑, we first employ a dialect-aware query reduction technique to derive a simplified query 𝑄𝑠 by removing 𝑑-irrelevant SQL elements from 𝑄𝑐. Subsequently, we utilize LLMs to translate 𝑄𝑠 into 𝑄𝑠′, and automatically extract the translation rule 𝑟𝑑 for dialect 𝑑 based on the relationship between 𝑄𝑠 and 𝑄𝑠′. By applying 𝑟𝑑 to 𝑄𝑐, we can effectively translate the dialect 𝑑 within 𝑄𝑐, thereby bypassing the complexity of the source query 𝑄𝑐. We evaluate RISE on two real-world benchmarks, i.e., TPC-DS and SQLProcBench, comparing its performance against both the traditional rule-based tools and the LLM-based approaches with respect to translation accuracy. RISE achieves accuracies of 97.98% on TPC-DS and 100% on SQLProcBench, outperforming the baselines by an average improvement of 24.62% and 238.41%, respectively.