Step-unrolled Denoising Autoencoders for Text Generation
Abstract
In this paper we propose a new generative model of text, Step-unrolled Denoising Autoencoder (SUNDAE), that does not rely on autoregressive models. Similarly to denoising diffusion techniques, SUNDAE is repeatedly applied on a sequence of tokens, starting from random inputs and improving them each time until convergence. We present a simple new improvement operator that converges in fewer iterations than diffusion methods, while qualitatively producing better samples on natural language datasets. SUNDAE achieves state-of-the-art results (among non-autoregressive methods) on the WMT'14 English-to-German translation task and good qualitative results on unconditional language modeling on the Colossal Cleaned Common Crawl dataset and a dataset of Python code from GitHub. The non-autoregressive nature of SUNDAE opens up possibilities beyond left-to-right prompted generation, by filling in arbitrary blank patterns in a template.
Keywords
Cite
@article{arxiv.2112.06749,
title = {Step-unrolled Denoising Autoencoders for Text Generation},
author = {Nikolay Savinov and Junyoung Chung and Mikolaj Binkowski and Erich Elsen and Aaron van den Oord},
journal= {arXiv preprint arXiv:2112.06749},
year = {2022}
}
Comments
Accepted to ICLR 2022