English

Focused Prefix Tuning for Controllable Text Generation

Computation and Language 2023-06-13 v2

Abstract

In a controllable text generation dataset, there exist unannotated attributes that could provide irrelevant learning signals to models that use it for training and thus degrade their performance. We propose focused prefix tuning(FPT) to mitigate the problem and to enable the control to focus on the desired attribute. Experimental results show that FPT can achieve better control accuracy and text fluency than baseline models in single-attribute control tasks. In multi-attribute control tasks, FPT achieves comparable control accuracy with the state-of-the-art approach while keeping the flexibility to control new attributes without retraining existing models.

Keywords

Cite

@article{arxiv.2306.00369,
  title  = {Focused Prefix Tuning for Controllable Text Generation},
  author = {Congda Ma and Tianyu Zhao and Makoto Shing and Kei Sawada and Manabu Okumura},
  journal= {arXiv preprint arXiv:2306.00369},
  year   = {2023}
}

Comments

Accepted to the ACL 2023

R2 v1 2026-06-28T10:52:54.245Z