Feather-SQL: A Lightweight NL2SQL Framework with Dual-Model Collaboration Paradigm for Small Language Models
This work addresses data privacy and deployment challenges in NL2SQL for resource-constrained environments, though it is incremental as it builds on existing SLM and collaboration concepts.
The paper tackles the problem of poor NL2SQL performance in small language models (SLMs) by introducing Feather-SQL, a lightweight framework with a dual-model collaboration paradigm, resulting in a 10% performance boost for non-fine-tuned SLMs and raising accuracy to 54.76% on the BIRD benchmark.
Natural Language to SQL (NL2SQL) has seen significant advancements with large language models (LLMs). However, these models often depend on closed-source systems and high computational resources, posing challenges in data privacy and deployment. In contrast, small language models (SLMs) struggle with NL2SQL tasks, exhibiting poor performance and incompatibility with existing frameworks. To address these issues, we introduce Feather-SQL, a new lightweight framework tailored for SLMs. Feather-SQL improves SQL executability and accuracy through 1) schema pruning and linking, 2) multi-path and multi-candidate generation. Additionally, we introduce the 1+1 Model Collaboration Paradigm, which pairs a strong general-purpose chat model with a fine-tuned SQL specialist, combining strong analytical reasoning with high-precision SQL generation. Experimental results on BIRD demonstrate that Feather-SQL improves NL2SQL performance on SLMs, with around 10% boost for models without fine-tuning. The proposed paradigm raises the accuracy ceiling of SLMs to 54.76%, highlighting its effectiveness.