Discrete Diffusion Modeling by Estimating the Ratios of the Data Distribution
This work addresses a key bottleneck in generative modeling for discrete domains, offering significant performance gains for language tasks, though it is incremental as it builds on existing diffusion frameworks.
The authors tackled the problem of applying diffusion models to discrete data like natural language by proposing score entropy, a novel loss that extends score matching to discrete spaces, resulting in SEDD models that reduce perplexity by 25-75% compared to existing language diffusion methods and outperform GPT-2 in some aspects.
Despite their groundbreaking performance for many generative modeling tasks, diffusion models have fallen short on discrete data domains such as natural language. Crucially, standard diffusion models rely on the well-established theory of score matching, but efforts to generalize this to discrete structures have not yielded the same empirical gains. In this work, we bridge this gap by proposing score entropy, a novel loss that naturally extends score matching to discrete spaces, integrates seamlessly to build discrete diffusion models, and significantly boosts performance. Experimentally, we test our Score Entropy Discrete Diffusion models (SEDD) on standard language modeling tasks. For comparable model sizes, SEDD beats existing language diffusion paradigms (reducing perplexity by $25$-$75$\%) and is competitive with autoregressive models, in particular outperforming GPT-2. Furthermore, compared to autoregressive mdoels, SEDD generates faithful text without requiring distribution annealing techniques like temperature scaling (around $6$-$8\times$ better generative perplexity than un-annealed GPT-2), can trade compute and quality (similar quality with $32\times$ fewer network evaluations), and enables controllable infilling (matching nucleus sampling quality while enabling other strategies besides left to right prompting).