Elucidating the design space of language models for image generation

TL;DR

This paper explores the design space of language models for image generation, analyzing their behavior, challenges, and scalability, and introduces ELM, which achieves state-of-the-art results on ImageNet 256x256.

Contribution

It provides the first comprehensive analysis of language model optimization in vision tasks and offers insights into model design choices for image generation.

Findings

Larger models better capture global image context.

Image tokens exhibit more randomness than text tokens.

ELM achieves state-of-the-art performance on ImageNet 256x256.

Abstract

The success of autoregressive (AR) language models in text generation has inspired the computer vision community to adopt Large Language Models (LLMs) for image generation. However, considering the essential differences between text and image modalities, the design space of language models for image generation remains underexplored. We observe that image tokens exhibit greater randomness compared to text tokens, which presents challenges when training with token prediction. Nevertheless, AR models demonstrate their potential by effectively learning patterns even from a seemingly suboptimal optimization problem. Our analysis also reveals that while all models successfully grasp the importance of local information in image generation, smaller models struggle to capture the global context. In contrast, larger models showcase improved capabilities in this area, helping to explain the performance gains achieved when scaling up model size. We further elucidate the design space of language models for vision generation, including tokenizer choice, model choice, model scalability, vocabulary design, and sampling strategy through extensive comparative experiments. Our work is the first to analyze the optimization behavior of language models in vision generation, and we believe it can inspire more effective designs when applying LMs to other domains. Finally, our elucidated language model for image generation, termed as ELM, achieves state-of-the-art performance on the ImageNet 256*256 benchmark. The code is available at https://github.com/Pepperlll/LMforImageGeneration.git.