Group Diffusion Transformers are Unsupervised Multitask Learners

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Huang, Lianghua, Wang, Wei, Wu, Zhi-Fan, Dou, Huanzhang, Shi, Yupeng, Feng, Yutong, Liang, Chen, Liu, Yu, Zhou, Jingren
Format: Preprint
Published: 2024
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866917809832853504
author Huang, Lianghua
Wang, Wei
Wu, Zhi-Fan
Dou, Huanzhang
Shi, Yupeng
Feng, Yutong
Liang, Chen
Liu, Yu
Zhou, Jingren
author_facet Huang, Lianghua
Wang, Wei
Wu, Zhi-Fan
Dou, Huanzhang
Shi, Yupeng
Feng, Yutong
Liang, Chen
Liu, Yu
Zhou, Jingren
contents While large language models (LLMs) have revolutionized natural language processing with their task-agnostic capabilities, visual generation tasks such as image translation, style transfer, and character customization still rely heavily on supervised, task-specific datasets. In this work, we introduce Group Diffusion Transformers (GDTs), a novel framework that unifies diverse visual generation tasks by redefining them as a group generation problem. In this approach, a set of related images is generated simultaneously, optionally conditioned on a subset of the group. GDTs build upon diffusion transformers with minimal architectural modifications by concatenating self-attention tokens across images. This allows the model to implicitly capture cross-image relationships (e.g., identities, styles, layouts, surroundings, and color schemes) through caption-based correlations. Our design enables scalable, unsupervised, and task-agnostic pretraining using extensive collections of image groups sourced from multimodal internet articles, image galleries, and video frames. We evaluate GDTs on a comprehensive benchmark featuring over 200 instructions across 30 distinct visual generation tasks, including picture book creation, font design, style transfer, sketching, colorization, drawing sequence generation, and character customization. Our models achieve competitive zero-shot performance without any additional fine-tuning or gradient updates. Furthermore, ablation studies confirm the effectiveness of key components such as data scaling, group size, and model design. These results demonstrate the potential of GDTs as scalable, general-purpose visual generation systems.
format Preprint
id arxiv_https___arxiv_org_abs_2410_15027
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Group Diffusion Transformers are Unsupervised Multitask Learners
Huang, Lianghua
Wang, Wei
Wu, Zhi-Fan
Dou, Huanzhang
Shi, Yupeng
Feng, Yutong
Liang, Chen
Liu, Yu
Zhou, Jingren
Computer Vision and Pattern Recognition
While large language models (LLMs) have revolutionized natural language processing with their task-agnostic capabilities, visual generation tasks such as image translation, style transfer, and character customization still rely heavily on supervised, task-specific datasets. In this work, we introduce Group Diffusion Transformers (GDTs), a novel framework that unifies diverse visual generation tasks by redefining them as a group generation problem. In this approach, a set of related images is generated simultaneously, optionally conditioned on a subset of the group. GDTs build upon diffusion transformers with minimal architectural modifications by concatenating self-attention tokens across images. This allows the model to implicitly capture cross-image relationships (e.g., identities, styles, layouts, surroundings, and color schemes) through caption-based correlations. Our design enables scalable, unsupervised, and task-agnostic pretraining using extensive collections of image groups sourced from multimodal internet articles, image galleries, and video frames. We evaluate GDTs on a comprehensive benchmark featuring over 200 instructions across 30 distinct visual generation tasks, including picture book creation, font design, style transfer, sketching, colorization, drawing sequence generation, and character customization. Our models achieve competitive zero-shot performance without any additional fine-tuning or gradient updates. Furthermore, ablation studies confirm the effectiveness of key components such as data scaling, group size, and model design. These results demonstrate the potential of GDTs as scalable, general-purpose visual generation systems.
title Group Diffusion Transformers are Unsupervised Multitask Learners
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2410.15027