Abstract
Controlling the behavior of language models (LMs) without re-training is a major
open problem in natural language generation. While recent works have demon-
strated successes on controlling simple sentence attributes (e.g., sentiment), there
has been little progress on complex, fine-grained controls (e.g., syntactic structure).
To address this challenge, we develop a new non-autoregressive language model
based on continuous diffusions that we call Diffusion-LM. Building upon the recent
successes of diffusion models in continuous domains, Diffusion-LM iteratively
denoises a sequence of Gaussian vectors into word vectors, yielding a sequence
of intermediate latent variables. The continuous, hierarchical nature of these inter-
mediate variables enables a simple gradient-based algorithm to perform complex,
controllable generation tasks. We demonstrate successful control of Diffusion-LM
for six challenging fine-grained control tasks, significantly outperforming prior
work
Users
Please
log in to take part in the discussion (add own reviews or comments).