A Cheaper and Better Diffusion Language Model with Soft-Masked Noise

Jiaao Chen, Aston Zhang, Mu Li, Alex Smola, Diyi Yang


Abstract
Diffusion models that are based on iterative denoising have been recently proposed and leveraged in various generation tasks like image generation. Whereas, as a way inherently built for continuous data, existing diffusion models still have some limitations in modeling discrete data, e.g., languages. For example, the generally used Gaussian noise can not handle the discrete corruption well, and the objectives in continuous spaces fail to be stable for textual data in the diffusion process especially when the dimension is high. To alleviate these issues, we introduce a novel diffusion model for language modeling, Masked-Diffuse LM, with lower training cost and better performances, inspired by linguistic features in languages. Specifically, we design a linguistic-informed forward process which adds corruptions to the text through strategically soft-masking to better noise the textual data. Also, we directly predict the categorical distribution with cross-entropy loss function in every diffusion step to connect the continuous space and discrete space in a more efficient and straightforward way. Through experiments on 5 controlled generation tasks, we demonstrate that our Masked-Diffuse LM can achieve better generation quality than the state-of-the-art diffusion models with better efficiency.
Anthology ID:
2023.emnlp-main.289
Volume:
Proceedings of the 2023 Conference on Empirical Methods in Natural Language Processing
Month:
December
Year:
2023
Address:
Singapore
Editors:
Houda Bouamor, Juan Pino, Kalika Bali
Venue:
EMNLP
SIG:
Publisher:
Association for Computational Linguistics
Note:
Pages:
4765–4775
Language:
URL:
https://aclanthology.org/2023.emnlp-main.289
DOI:
10.18653/v1/2023.emnlp-main.289
Bibkey:
Cite (ACL):
Jiaao Chen, Aston Zhang, Mu Li, Alex Smola, and Diyi Yang. 2023. A Cheaper and Better Diffusion Language Model with Soft-Masked Noise. In Proceedings of the 2023 Conference on Empirical Methods in Natural Language Processing, pages 4765–4775, Singapore. Association for Computational Linguistics.
Cite (Informal):
A Cheaper and Better Diffusion Language Model with Soft-Masked Noise (Chen et al., EMNLP 2023)
Copy Citation:
PDF:
https://aclanthology.org/2023.emnlp-main.289.pdf
Video:
 https://aclanthology.org/2023.emnlp-main.289.mp4