SVG-T2I: Scaling Up Text-to-Image Latent Diffusion Model Without Variational Autoencoder
This work addresses the problem of integrating visual understanding and generation for AI researchers, though it is incremental as it scales an existing framework.
The paper tackles training text-to-image diffusion models directly in Visual Foundation Model (VFM) representation space, achieving competitive results with scores of 0.75 on GenEval and 85.78 on DPG-Bench.
Visual generation grounded in Visual Foundation Model (VFM) representations offers a highly promising unified pathway for integrating visual understanding, perception, and generation. Despite this potential, training large-scale text-to-image diffusion models entirely within the VFM representation space remains largely unexplored. To bridge this gap, we scale the SVG (Self-supervised representations for Visual Generation) framework, proposing SVG-T2I to support high-quality text-to-image synthesis directly in the VFM feature domain. By leveraging a standard text-to-image diffusion pipeline, SVG-T2I achieves competitive performance, reaching 0.75 on GenEval and 85.78 on DPG-Bench. This performance validates the intrinsic representational power of VFMs for generative tasks. We fully open-source the project, including the autoencoder and generation model, together with their training, inference, evaluation pipelines, and pre-trained weights, to facilitate further research in representation-driven visual generation.