Accelerating Masked Image Generation by Learning Latent Controlled Dynamics

TL;DR

This paper introduces MIGM-Shortcut, a lightweight model that accelerates masked image generation by learning feature dynamics, achieving over 4x faster text-to-image synthesis without quality loss.

Contribution

It proposes a novel method to accelerate masked image generation by learning latent feature dynamics, reducing computation while maintaining high quality.

Findings

Over 4x acceleration in text-to-image generation.

Maintains image quality comparable to original models.

Applicable to multiple architectures and tasks.

Abstract

Masked Image Generation Models (MIGMs) have achieved great success, yet their efficiency is hampered by the multiple steps of bi-directional attention. In fact, there exists notable redundancy in their computation: when sampling discrete tokens, the rich semantics contained in the continuous features are lost. Some existing works attempt to cache the features to approximate future features. However, they exhibit considerable approximation error under aggressive acceleration rates. We attribute this to their limited expressivity and the failure to account for sampling information. To fill this gap, we propose to learn a lightweight model that incorporates both previous features and sampled tokens, and regresses the average velocity field of feature evolution. The model has moderate complexity that suffices to capture the subtle dynamics while keeping lightweight compared to the original base model. We apply our method, MIGM-Shortcut, to two representative MIGM architectures and tasks. In particular, on the state-of-the-art Lumina-DiMOO, it achieves over 4x acceleration of text-to-image generation while maintaining quality, significantly pushing the Pareto frontier of masked image generation. The code and model weights are available at https://github.com/Kaiwen-Zhu/MIGM-Shortcut.