Remasking Discrete Diffusion Models with Inference-Time Scaling

Guanghan Wang, Yair Schiff, Subham Sekhar Sahoo, Volodymyr Kuleshov

Keywords: discrete diffusion, diffusion language models, diffusion models, diffusion samplers, probabilistic generative models

TL;DR: Absorbing state discrete diffusion models that enable remasking benefit from inference-time scaling, improving sample quality, controlled generation, and performance on downstream tasks.

Abstract: Part of the success of diffusion models stems from their ability to perform iterative refinement, i.e., repeatedly correcting outputs during generation. However, modern masked discrete diffusion lacks this capability: when a token is generated, it cannot be updated again, even when it introduces an error. Here, we address this limitation by introducing the remasking diffusion model (ReMDM) sampler, a method that can be applied to pretrained masked diffusion models in a principled way and that is derived from a discrete diffusion model with a custom remasking backward process. Most interestingly, ReMDM endows discrete diffusion with a form of inference-time compute scaling. By increasing the number of sampling steps, ReMDM generates natural language outputs that approach the quality of autoregressive models, whereas when the computation budget is limited, ReMDM better maintains quality. ReMDM also improves sample quality of masked diffusion models for discretized images, and in scientific domains such as molecule design, ReMDM facilitates diffusion guidance and pushes the Pareto frontier of controllability relative to classical masking and uniform noise diffusion. When applied to large pretrained diffusion language models, ReMDM boosts the model’s performance on downstream tasks requiring factual knowledge grasp and reasoning ability.

Primary Area: Deep learning (e.g., architectures, generative models, optimization for deep networks, foundation models, LLMs)

Submission Number: 23437