acceptodds
Desk rejected at ICLR 2027

NesTok: Nested Self-Aligned 1D Tokenizer for Autoregressive Image Generation

Desk rejected at ICLR 2027, so left out of the paper lists and the map.

Abstract

One-dimensional (1D) variable-length visual tokenizers enable adaptive compression by varying the number of tokens, allowing downstream autoregressive (AR) models to flexibly trade off generation quality against computational cost using a single tokenizer. However, existing approaches based on nested dropout often fail to fully exploit the representational capacity of the tokenizer, resulting in suboptimal performance in both image reconstruction and generation. In this work, we introduce NesTok, a nested self-alignment framework tailored to dynamic visual tokenizers. NesTok introduces cross-length training, which jointly optimizes reconstruction across token lengths while using the full-length sequence to guide shorter counterparts, enabling shorter token sequences to approach the reconstruction quality of full-length sequences. On ImageNet, NesTok improves substantially over standard training and achieves an rFID score of 0.98. On downstream image generation, it achieves the state-of-the-art gFID score of on Imagenet 256256 among existing variable-length autoregressive image generation methods.

Primary area:
generative models
Keywords:
Image generation, Autoregressive models, Tokenizer

Related papers

Discussion (0)

Sign in to comment.