Why Masking Diffusion Works: Condition on the Jump Schedule for Improved Discrete Diffusion

• URL: http://arxiv.org/abs/2506.08316v1
• Date: Tue, 10 Jun 2025 00:58:25 GMT
• Title: Why Masking Diffusion Works: Condition on the Jump Schedule for Improved Discrete Diffusion
• Authors: Alan N. Amin, Nate Gruver, Andrew Gordon Wilson,
• Abstract summary: Markov processes evolve by discontinuous jumps at a fixed rate.<n>Unlike other discrete diffusion models, masking diffusion builds in the known distribution of jump times and only learns where to jump to.
• Score: 45.651372465763615
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Discrete diffusion models, like continuous diffusion models, generate high-quality samples by gradually undoing noise applied to datapoints with a Markov process. Gradual generation in theory comes with many conceptual benefits; for example, inductive biases can be incorporated into the noising Markov process, and access to improved sampling algorithms. In practice, however, the consistently best performing discrete diffusion model is, surprisingly, masking diffusion, which does not denoise gradually. Here we explain the superior performance of masking diffusion by noting that it makes use of a fundamental difference between continuous and discrete Markov processes: discrete Markov processes evolve by discontinuous jumps at a fixed rate and, unlike other discrete diffusion models, masking diffusion builds in the known distribution of jump times and only learns where to jump to. We show that we can similarly bake in the known distribution of jump times into any discrete diffusion model. The resulting models - schedule-conditioned discrete diffusion (SCUD) - generalize classical discrete diffusion and masking diffusion. By applying SCUD to models with noising processes that incorporate inductive biases on images, text, and protein data, we build models that outperform masking.

Related papers

• Generative modelling with jump-diffusions [0.0]
  I present a generalization of generative diffusion processes to a wide class of non-Gaussian noise processes.<n>For the problem of capturing a heavy-tailed target distribution, the jump-diffusion Laplace model outperforms models driven by alpha-stable noise.
  arXiv  Detail & Related papers  (2025-03-09T11:08:03Z)
• Generalized Interpolating Discrete Diffusion [65.74168524007484]
  Masked diffusion is a popular choice due to its simplicity and effectiveness.<n>We generalize a new family of general interpolating discrete diffusion (GIDD) which offers greater flexibility in the design of the noising processes.<n>Exploiting GIDD's flexibility, we explore a hybrid approach combining masking and uniform noise, leading to improved sample quality.
  arXiv  Detail & Related papers  (2025-03-06T14:30:55Z)
• Continuous Diffusion Model for Language Modeling [57.396578974401734]
  Existing continuous diffusion models for discrete data have limited performance compared to discrete approaches.<n>We propose a continuous diffusion model for language modeling that incorporates the geometry of the underlying categorical distribution.
  arXiv  Detail & Related papers  (2025-02-17T08:54:29Z)
• Convergence Analysis of Discrete Diffusion Model: Exact Implementation through Uniformization [17.535229185525353]
  We introduce an algorithm leveraging the uniformization of continuous Markov chains, implementing transitions on random time points. Our results align with state-of-the-art achievements for diffusion models in $mathbbRd$ and further underscore the advantages of discrete diffusion models in comparison to the $mathbbRd$ setting.
  arXiv  Detail & Related papers  (2024-02-12T22:26:52Z)
• Guided Diffusion from Self-Supervised Diffusion Features [49.78673164423208]
  Guidance serves as a key concept in diffusion models, yet its effectiveness is often limited by the need for extra data annotation or pretraining. We propose a framework to extract guidance from, and specifically for, diffusion models.
  arXiv  Detail & Related papers  (2023-12-14T11:19:11Z)
• Generative Fractional Diffusion Models [53.36835573822926]
  We introduce the first continuous-time score-based generative model that leverages fractional diffusion processes for its underlying dynamics. Our evaluations on real image datasets demonstrate that GFDM achieves greater pixel-wise diversity and enhanced image quality, as indicated by a lower FID.
  arXiv  Detail & Related papers  (2023-10-26T17:53:24Z)
• Soft Mixture Denoising: Beyond the Expressive Bottleneck of Diffusion Models [76.46246743508651]
  We show that current diffusion models actually have an expressive bottleneck in backward denoising. We introduce soft mixture denoising (SMD), an expressive and efficient model for backward denoising.
  arXiv  Detail & Related papers  (2023-09-25T12:03:32Z)
• Blackout Diffusion: Generative Diffusion Models in Discrete-State Spaces [0.0]
  We develop a theoretical formulation for arbitrary discrete-state Markov processes in the forward diffusion process. As an example, we introduce Blackout Diffusion'', which learns to produce samples from an empty image instead of from noise.
  arXiv  Detail & Related papers  (2023-05-18T16:24:12Z)
• Unifying Diffusion Models' Latent Space, with Applications to CycleDiffusion and Guidance [95.12230117950232]
  We show that a common latent space emerges from two diffusion models trained independently on related domains. Applying CycleDiffusion to text-to-image diffusion models, we show that large-scale text-to-image diffusion models can be used as zero-shot image-to-image editors.
  arXiv  Detail & Related papers  (2022-10-11T15:53:52Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.