Discrete Flow Matching

• URL: http://arxiv.org/abs/2407.15595v2
• Date: Tue, 5 Nov 2024 10:02:42 GMT
• Title: Discrete Flow Matching
• Authors: Itai Gat, Tal Remez, Neta Shaul, Felix Kreuk, Ricky T. Q. Chen, Gabriel Synnaeve, Yossi Adi, Yaron Lipman,
• Abstract summary: We present a novel discrete flow paradigm designed specifically for generating discrete data. Our approach is capable of generating high-quality discrete data in a non-autoregressive fashion.
• Score: 74.04153927689313
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Despite Flow Matching and diffusion models having emerged as powerful generative paradigms for continuous variables such as images and videos, their application to high-dimensional discrete data, such as language, is still limited. In this work, we present Discrete Flow Matching, a novel discrete flow paradigm designed specifically for generating discrete data. Discrete Flow Matching offers several key contributions:(i) it works with a general family of probability paths interpolating between source and target distributions; (ii) it allows for a generic formula for sampling from these probability paths using learned posteriors such as the probability denoiser ($x$-prediction) and noise-prediction ($\epsilon$-prediction); (iii) practically, focusing on specific probability paths defined with different schedulers improves generative perplexity compared to previous discrete diffusion and flow models; and (iv) by scaling Discrete Flow Matching models up to 1.7B parameters, we reach 6.7% Pass@1 and 13.4% Pass@10 on HumanEval and 6.7% Pass@1 and 20.6% Pass@10 on 1-shot MBPP coding benchmarks. Our approach is capable of generating high-quality discrete data in a non-autoregressive fashion, significantly closing the gap between autoregressive models and discrete flow models.

Related papers

• Diffusion Forcing: Next-token Prediction Meets Full-Sequence Diffusion [61.03681839276652]
  Diffusion Forcing is a new training paradigm where a diffusion model is trained to denoise a set of tokens with independent per-token noise levels. We apply Diffusion Forcing to sequence generative modeling by training a causal next-token prediction model to generate one or several future tokens.
  arXiv  Detail & Related papers  (2024-07-01T15:43:25Z)
• Bayesian Flow Networks [4.585102332532472]
  This paper introduces Bayesian Flow Networks (BFNs), a new class of generative model in which the parameters of a set of independent distributions are modified with Bayesian inference. Starting from a simple prior and iteratively updating the two distributions yields a generative procedure similar to the reverse process of diffusion models. BFNs achieve competitive log-likelihoods for image modelling on dynamically binarized MNIST and CIFAR-10, and outperform all known discrete diffusion models on the text8 character-level language modelling task.
  arXiv  Detail & Related papers  (2023-08-14T09:56:35Z)
• Towards Faster Non-Asymptotic Convergence for Diffusion-Based Generative Models [49.81937966106691]
  We develop a suite of non-asymptotic theory towards understanding the data generation process of diffusion models. In contrast to prior works, our theory is developed based on an elementary yet versatile non-asymptotic approach.
  arXiv  Detail & Related papers  (2023-06-15T16:30:08Z)
• The Score-Difference Flow for Implicit Generative Modeling [1.309716118537215]
  Implicit generative modeling aims to produce samples of synthetic data matching a target data distribution. Recent work has approached the IGM problem from the perspective of pushing synthetic source data toward the target distribution. We present the score difference between arbitrary target and source distributions as a flow that optimally reduces the Kullback-Leibler divergence between them.
  arXiv  Detail & Related papers  (2023-04-25T15:21:12Z)
• Stochastic Interpolants: A Unifying Framework for Flows and Diffusions [16.95541777254722]
  A class of generative models that unifies flow-based and diffusion-based methods is introduced. These models extend the framework proposed in Albergo & VandenEijnden (2023), enabling the use of a broad class of continuous-time processes called stochastic interpolants' These interpolants are built by combining data from the two prescribed densities with an additional latent variable that shapes the bridge in a flexible way.
  arXiv  Detail & Related papers  (2023-03-15T17:43:42Z)
• Score-based Continuous-time Discrete Diffusion Models [102.65769839899315]
  We extend diffusion models to discrete variables by introducing a Markov jump process where the reverse process denoises via a continuous-time Markov chain. We show that an unbiased estimator can be obtained via simple matching the conditional marginal distributions. We demonstrate the effectiveness of the proposed method on a set of synthetic and real-world music and image benchmarks.
  arXiv  Detail & Related papers  (2022-11-30T05:33:29Z)
• Matching Normalizing Flows and Probability Paths on Manifolds [57.95251557443005]
  Continuous Normalizing Flows (CNFs) are generative models that transform a prior distribution to a model distribution by solving an ordinary differential equation (ODE) We propose to train CNFs by minimizing probability path divergence (PPD), a novel family of divergences between the probability density path generated by the CNF and a target probability density path. We show that CNFs learned by minimizing PPD achieve state-of-the-art results in likelihoods and sample quality on existing low-dimensional manifold benchmarks.
  arXiv  Detail & Related papers  (2022-07-11T08:50:19Z)
• Sparse Communication via Mixed Distributions [29.170302047339174]
  We build theoretical foundations for "mixed random variables" Our framework suggests two strategies for representing and sampling mixed random variables. We experiment with both approaches on an emergent communication benchmark.
  arXiv  Detail & Related papers  (2021-08-05T14:49:03Z)
• Discrete Denoising Flows [87.44537620217673]
  We introduce a new discrete flow-based model for categorical random variables: Discrete Denoising Flows (DDFs) In contrast with other discrete flow-based models, our model can be locally trained without introducing gradient bias. We show that DDFs outperform Discrete Flows on modeling a toy example, binary MNIST and Cityscapes segmentation maps, measured in log-likelihood.
  arXiv  Detail & Related papers  (2021-07-24T14:47:22Z)
• Variational Mixture of Normalizing Flows [0.0]
  Deep generative models, such as generative adversarial networks autociteGAN, variational autoencoders autocitevaepaper, and their variants, have seen wide adoption for the task of modelling complex data distributions. Normalizing flows have overcome this limitation by leveraging the change-of-suchs formula for probability density functions. The present work overcomes this by using normalizing flows as components in a mixture model and devising an end-to-end training procedure for such a model.
  arXiv  Detail & Related papers  (2020-09-01T17:20:08Z)

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.