Heptapod: Language Modeling on Visual Signals
This work addresses image generation for computer vision researchers, offering a novel approach that combines autoregressive and masked autoencoding principles, though it appears incremental in advancing existing paradigms.
The paper tackles image generation by introducing Heptapod, an autoregressive model that predicts the next 2D distribution of images, achieving an FID of 2.70 on ImageNet, which significantly outperforms prior causal autoregressive methods.
We introduce Heptapod, an image autoregressive model that adheres to the foundational principles of language modeling. Heptapod employs \textbf{causal attention}, \textbf{eliminates reliance on CFG}, and \textbf{eschews the trend of semantic tokenizers}. Our key innovation is \textit{next 2D distribution prediction}: a causal Transformer with reconstruction-focused visual tokenizer, learns to predict the distribution over the entire 2D spatial grid of images at each timestep. This learning objective unifies the sequential modeling of autoregressive framework with the holistic self-supervised learning of masked autoencoding, enabling the model to capture comprehensive image semantics via generative training. On the ImageNet generation benchmark, Heptapod achieves an FID of $2.70$, significantly outperforming previous causal autoregressive approaches. We hope our work inspires a principled rethinking of language modeling on visual signals and beyond.