Group Diffusion Transformers are Unsupervised Multitask Learners
This addresses the need for scalable, general-purpose visual generation systems for applications like image translation and style transfer, though it builds incrementally on diffusion transformers.
The paper tackles the problem of visual generation tasks requiring supervised, task-specific datasets by introducing Group Diffusion Transformers (GDTs), which unify diverse tasks as group generation problems and achieve competitive zero-shot performance on over 200 instructions across 30 tasks without fine-tuning.
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.