vix.ing · top · new · best · stats

Diffusion-LM Improves Controllable Text Generation

2022/05/27 by Xiang Lisa Li, John Thickstun, Li, Xiang Lisa +8 · 226 citations
Computer Science · #Artificial Intelligence (cs.AI) #Computation and Language (cs.CL) #FOS: Computer and information sciences #Machine Learning (cs.LG) #Natural Language Processing Techniques #Speech Recognition and Synthesis #Topic Modeling #cs.AI #cs.CL #cs.LG

paper · pdf · doi:10.48550/arxiv.2205.14217

arxiv created 2022/05/27 · openalex publication_date 2022/05/27 · arxiv updated 2022/05/31 · openalex created_date 2025/10/10 · openalex updated_date 2026/07/28

Abstract

Controlling the behavior of language models (LMs) without re-training is a major open problem in natural language generation. While recent works have demonstrated 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 intermediate 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.

Cited by

Related