0

Implicit Unlikelihood Training: Improving Neural Text Generation with Reinforcement Learning

Fine-tuning a language model using policy gradient reinforcement learning and unlikelihood training improves text generation by reducing repetition without sacrificing quality.

Year
2021
Venue
EACL 2021 2
Authors
3
Hosting
Abstract onlyARXIV-DEFAULT

Cite

Notes

Only stored in your browser.

Attribution

Abstract & full text
arxiv.org/abs/2101.04229ARXIV-DEFAULT
TL;DR
Semantic Scholar
Attribution policy →

Abstract

Likelihood training and maximization-based decoding result in dull and repetitive generated texts even when using powerful language models (Holtzman et al., 2019). Adding a loss function for regularization was shown to improve text generation output by helping avoid unwanted properties, such as contradiction or repetition (Li at al., 2020). In this work, we propose fine-tuning a language model by using policy gradient reinforcement learning, directly optimizing for better generation. We apply this approach to minimizing repetition in generated text, and show that, when combined with unlikelihood training (Welleck et al., 2020), our method further reduces repetition without impacting the language model quality. We also evaluate other methods for improving generation at training and decoding time, and compare them using various metrics aimed at control for better text generation output.

Authors

3