no code implementations • 27 May 2024 • Shayan Talaei, Mohammadreza Pourreza, Yu-Chen Chang, Azalia Mirhoseini, Amin Saberi
Additionally, we have developed an adaptive schema pruning technique that adjusts based on the complexity of the problem and the model's context size.
no code implementations • 24 Mar 2024 • Mohammadreza Pourreza, Davood Rafiei, Yuxi Feng, Raymond Li, Zhenan Fan, Weiwei Zhang
Furthermore, compared to these competitive models, our proposed encoder enhances the downstream performance of NL2SQL models in 1-shot in-context learning scenarios by 1-2\% for GPT-3. 5-turbo, 4-8\% for CodeLlama-7B, and 2-3\% for CodeLlama-13B.
no code implementations • 2 Feb 2024 • Mohammadreza Pourreza, Davood Rafiei
Leading models for the text-to-SQL task heavily rely on proprietary Large Language Models (LLMs), posing concerns over data privacy.
no code implementations • 27 Oct 2023 • Mohammadreza Pourreza, Davood Rafiei
In this paper, we conduct an extensive study of several prominent cross-domain text-to-SQL benchmarks and re-evaluate some of the top-performing models within these benchmarks, by both manually evaluating the SQL queries and rewriting them in equivalent expressions.
1 code implementation • NeurIPS 2023 • Mohammadreza Pourreza, Davood Rafiei
In particular, we show that breaking down the generation problem into sub-problems and feeding the solutions of those sub-problems into LLMs can be an effective approach for significantly improving their performance.
Ranked #3 on Text-To-SQL on spider