Distilling Text Style Transfer With Self-Explanation From LLMs

arXiv:2403.01106v1 Announce Type: new
Abstract: Text Style Transfer (TST) seeks to alter the style of text while retaining its core content. Given the constraints of limited parallel datasets for TST, we propose CoTeX, a framework that leverages large language models (LLMs) alongside chain-of-thought (CoT) prompting to facilitate TST. CoTeX distills the complex rewriting and reasoning capabilities of LLMs into more streamlined models capable of working with both non-parallel and parallel data. Through experimentation across four TST datasets, CoTeX is shown …

abstract arxiv constraints core cs.ai cs.cl datasets framework language language models large language large language models llms prompting style style transfer text text style transfer thought transfer type