Markup-to-Image Diffusion Models with Scheduled Sampling

TL;DR

This paper introduces a diffusion-based method with scheduled sampling for converting markup languages into images, improving generation quality across diverse datasets.

Contribution

It adapts scheduled sampling to diffusion models for markup-to-image tasks, addressing exposure bias and enhancing generation accuracy.

Findings

Effective diffusion process verified across datasets

Scheduled sampling mitigates generation errors

Markup-to-image task aids in analyzing generative models

Abstract

Building on recent advances in image generation, we present a fully data-driven approach to rendering markup into images. The approach is based on diffusion models, which parameterize the distribution of data using a sequence of denoising operations on top of a Gaussian noise distribution. We view the diffusion denoising process as a sequential decision making process, and show that it exhibits compounding errors similar to exposure bias issues in imitation learning problems. To mitigate these issues, we adapt the scheduled sampling algorithm to diffusion training. We conduct experiments on four markup datasets: mathematical formulas (LaTeX), table layouts (HTML), sheet music (LilyPond), and molecular images (SMILES). These experiments each verify the effectiveness of the diffusion process and the use of scheduled sampling to fix generation issues. These results also show that the markup-to-image task presents a useful controlled compositional setting for diagnosing and analyzing generative image models.