English

StyleDGPT: Stylized Response Generation with Pre-trained Language Models

Computation and Language 2020-10-07 v1

Abstract

Generating responses following a desired style has great potentials to extend applications of open-domain dialogue systems, yet is refrained by lacking of parallel data for training. In this work, we explore the challenging task with pre-trained language models that have brought breakthrough to various natural language tasks. To this end, we introduce a KL loss and a style classifier to the fine-tuning step in order to steer response generation towards the target style in both a word-level and a sentence-level. Comprehensive empirical studies with two public datasets indicate that our model can significantly outperform state-of-the-art methods in terms of both style consistency and contextual coherence.

Keywords

Cite

@article{arxiv.2010.02569,
  title  = {StyleDGPT: Stylized Response Generation with Pre-trained Language Models},
  author = {Ze Yang and Wei Wu and Can Xu and Xinnian Liang and Jiaqi Bai and Liran Wang and Wei Wang and Zhoujun Li},
  journal= {arXiv preprint arXiv:2010.02569},
  year   = {2020}
}

Comments

Findings of EMNLP2020

R2 v1 2026-06-23T19:04:45.660Z