Diffusion-LM Improves Controllable Text Generation

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 in corpus (10)

• A Diffusion model for POI recommendation
• Diffusion-based Contrastive Learning for Sequential Recommendation
• DiffDance: Cascaded Human Motion Diffusion Model for Dance Generation
• Diffusion-based Document Layout Generation
• DeTiME: Diffusion-Enhanced Topic Modeling using Encoder-decoder based LLM
• Score-Based Generative Models for PET Image Reconstruction
• 4D Facial Expression Diffusion Model
• Unraveling the Potential of Diffusion Models in Small Molecule Generation
• Table-to-Text Generation with Pretrained Diffusion Models
• ForceGen: End-to-end de novo protein generation based on nonlinear mechanical unfolding responses using a protein language diffusion model