ARLON: Boosting Diffusion Transformers With Autoregressive Models for Long Video Generation

Zongyi Li^1* Shujie Hu^2* Shujie Liu^3† Long Zhou³ Jeongsoo Choi⁴ Lingwei Meng³ Xun Guo³ Jinyu Li³ Hefei Ling¹ Furu Wei^3
1Huazhong University of Science and Technology ²The Chinese University of Hong Kong ³Microsoft Corporation ⁴KAIST
^*Equal contribution ^†Corresponding author

Abstract

Text-to-video (T2V) models have recently undergone rapid and substantial advancements. Nevertheless, due to limitations in data and computational resources, achieving efficient generation of long videos with rich motion dynamics remains a significant challenge. To generate high-quality, dynamic, and temporally consistent long videos, this paper presents ARLON, a novel framework that boosts diffusion Transformers with autoregressive (AR) models for long (LON) video generation, by integrating the coarse spatial and long-range temporal information provided by the AR model to guide the DiT model effectively. Specifically, ARLON incorporates several key innovations: 1) A latent Vector Quantized Variational Autoencoder (VQ-VAE) compresses the input latent space of the DiT model into compact and highly quantized visual tokens, bridging the AR and DiT models and balancing the learning complexity and information density; 2) An adaptive norm-based semantic injection module integrates the coarse discrete visual units from the AR model into the DiT model, ensuring effective guidance during video generation; 3) To enhance the tolerance capability of noise introduced from the AR inference, the DiT model is trained with coarser visual latent tokens incorporated with an uncertainty sampling module. Experimental results demonstrate that ARLON significantly outperforms the baseline OpenSora-V1.2 on eight out of eleven metrics selected from VBench, with notable improvements in dynamic degree and aesthetic quality, while delivering competitive results on the remaining three and simultaneously accelerating the generation process. In addition, ARLON achieves state-of-the-art performance in long video generation, outperforming other open-source models in this domain. Detailed analyses of the improvements in inference efficiency are presented, alongside a practical application that demonstrates the generation of long videos using progressive text prompts.

Method

Our method first generates long-term, coarse-grained discrete visual units (AR codes) autoregressively using a decoder-only Transformer. These discrete AR codes are then segmented and sequentially fed into the DiT model by the proposed semantic injection module, which autoregressively generates high-quality video segments.

Given a text prompt, the autoregressive (AR) model predicts coarse visual latent tokens, which are constructed from a 3D VAE encoder followed by a latent VQ-VAE encoder based on the target video. These predicted visual latent tokens encapsulate both the coarse spatial information and consistent semantic information. Based on these tokens, a latent VQ-VAE decoder generates continuous latent features, which serve as semantic conditions to guide the DiT model with a semantic injection module. To mitigate the noise inevitably introduced during AR inference, we introduce two noise-robust training strategies: 1) coarser visual latent tokens, and 2) uncertainty sampling module.

Denoising Step Results