English

Distilling Text Style Transfer With Self-Explanation From LLMs

Computation and Language 2024-05-07 v2 Artificial Intelligence

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 to surpass traditional supervised fine-tuning and knowledge distillation methods, particularly in low-resource settings. We conduct a comprehensive evaluation, comparing CoTeX against current unsupervised, supervised, in-context learning (ICL) techniques, and instruction-tuned LLMs. Furthermore, CoTeX distinguishes itself by offering transparent explanations for its style transfer process.

Keywords

Cite

@article{arxiv.2403.01106,
  title  = {Distilling Text Style Transfer With Self-Explanation From LLMs},
  author = {Chiyu Zhang and Honglong Cai and Yuezhang and Li and Yuexin Wu and Le Hou and Muhammad Abdul-Mageed},
  journal= {arXiv preprint arXiv:2403.01106},
  year   = {2024}
}

Comments

Accepted by NAACL Student Research Workshop 2024

R2 v1 2026-06-28T15:06:55.896Z