DiffusionBERT: Improving Generative Masked Language Models with Diffusion Models

Zhengfu He, Tianxiang Sun, Qiong Tang, Kuanning Wang, Xuanjing Huang, Xipeng Qiu


Abstract
We present DiffusionBERT, a new generative masked language model based on discrete dif- fusion models. Diffusion models and many pre- trained language models have a shared training objective, i.e., denoising, making it possible to combine the two powerful models and enjoy the best of both worlds. On the one hand, dif- fusion models offer a promising training strat- egy that helps improve the generation quality. On the other hand, pre-trained denoising lan- guage models (e.g., BERT) can be used as a good initialization that accelerates convergence. We explore training BERT to learn the reverse process of a discrete diffusion process with an absorbing state and elucidate several designs to improve it. First, we propose a new noise schedule for the forward diffusion process that controls the degree of noise added at each step based on the information of each token. Sec- ond, we investigate several designs of incorpo- rating the time step into BERT. Experiments on unconditional text generation demonstrate that DiffusionBERT achieves significant improve- ment over existing diffusion models for text (e.g., D3PM and Diffusion-LM) and previous generative masked language models in terms of perplexity and BLEU score. Promising re- sults in conditional generation tasks show that DiffusionBERT can generate texts of compa- rable quality and more diverse than a series of established baselines.
Anthology ID:
2023.acl-long.248
Volume:
Proceedings of the 61st Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)
Month:
July
Year:
2023
Address:
Toronto, Canada
Editors:
Anna Rogers, Jordan Boyd-Graber, Naoaki Okazaki
Venue:
ACL
SIG:
Publisher:
Association for Computational Linguistics
Note:
Pages:
4521–4534
Language:
URL:
https://aclanthology.org/2023.acl-long.248
DOI:
10.18653/v1/2023.acl-long.248
Bibkey:
Cite (ACL):
Zhengfu He, Tianxiang Sun, Qiong Tang, Kuanning Wang, Xuanjing Huang, and Xipeng Qiu. 2023. DiffusionBERT: Improving Generative Masked Language Models with Diffusion Models. In Proceedings of the 61st Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers), pages 4521–4534, Toronto, Canada. Association for Computational Linguistics.
Cite (Informal):
DiffusionBERT: Improving Generative Masked Language Models with Diffusion Models (He et al., ACL 2023)
Copy Citation:
PDF:
https://preview.aclanthology.org/ingest-acl-2023-videos/2023.acl-long.248.pdf