Improving Demonstration Diversity by Human-Free Fusing for Text-to-SQL

arXiv:2402.10663v1 Announce Type: new
Abstract: Currently, the in-context learning method based on large language models (LLMs) has become the mainstream of text-to-SQL research. Previous works have discussed how to select demonstrations related to the user question from a human-labeled demonstration pool. However, human labeling suffers from the limitations of insufficient diversity and high labeling overhead. Therefore, in this paper, we discuss how to measure and improve the diversity of the demonstrations for text-to-SQL. We present a metric to measure the …

abstract arxiv become context cs.cl diversity free human in-context learning labeling language language models large language large language models limitations llms pool question research sql text text-to-sql type