Commonly used image tokenizers produce a 2D grid of spatially arranged tokens. In contrast, so-called 1D image tokenizers represent images as highly compressed one-dimensional sequences of as few as 32 discrete tokens. We find that the high degree of compression achieved by a 1D tokenizer with vector quantization enables image editing and generative capabilities through heuristic manipulation of tokens, demonstrating that even very crude manipulations -- such as copying and replacing tokens between latent representations of images -- enable fine-grained image editing by transferring appearance and semantic attributes. Motivated by the expressivity of the 1D tokenizer's latent space, we construct an image generation pipeline leveraging gradient-based test-time optimization of tokens with plug-and-play loss functions such as reconstruction or CLIP similarity. Our approach is demonstrated for inpainting and text-guided image editing use cases, and can generate diverse and realistic samples without requiring training of any generative model.
@article{arxiv.2506.08257,
title = {Highly Compressed Tokenizer Can Generate Without Training},
author = {L. Lao Beyer and T. Li and X. Chen and S. Karaman and K. He},
journal= {arXiv preprint arXiv:2506.08257},
year = {2025}
}
Comments
Main manuscript: 9 pages, 7 figures. Appendix: 8 pages, 9 figures. To appear in the Proceedings of the 42nd International Conference on Machine Learning