Generative Pre-trained Autoregressive Diffusion Transformer

Yuan Zhang, Jiacheng Jiang, Guoqing Ma, Zhiying Lu, Bo Wang, Haoyang Huang, Jianlong Yuan, Nan Duan

Keywords: video generation, auto-regressive, diffusion

TL;DR: We develop a new framework that combines autoregressive modeling and diffusion for video data.

Abstract: In this work, we present GPDiT, a Generative Pre-trained Autoregressive Diffusion Transformer that unifies the strengths of diffusion and autoregressive modeling for long-range video synthesis, within a continuous latent space. Instead of predicting discrete tokens, GPDiT autoregressively predicts future latent frames using a diffusion loss, enabling natural modeling of motion dynamics and semantic consistency across frames. This continuous autoregressive framework not only enhances generation quality but also endows the model with representation capabilities. Additionally, we introduce a lightweight causal attention variant and a parameter-free rotation-based time-conditioning mechanism, improving both the training and inference efficiency. Extensive experiments demonstrate that GPDiT achieves strong performance in video generation quality, video representation ability, and few-shot learning tasks, highlighting its potential as an effective framework for video modeling in continuous space.

Primary Area: Deep learning (e.g., architectures, generative models, optimization for deep networks, foundation models, LLMs)

Flagged For Ethics Review: true

Submission Number: 15795