YaART: Yet Another ART Rendering Technology
This work addresses the need for efficient and high-quality text-to-image generation systems, which is incremental as it builds on existing cascaded diffusion models with a focus on dataset and model size optimization.
The study tackled the problem of developing efficient and high-fidelity text-to-image diffusion models by introducing YaART, a cascaded diffusion model aligned with human preferences using RLHF, which was consistently preferred by users over many existing state-of-the-art models.
In the rapidly progressing field of generative models, the development of efficient and high-fidelity text-to-image diffusion systems represents a significant frontier. This study introduces YaART, a novel production-grade text-to-image cascaded diffusion model aligned to human preferences using Reinforcement Learning from Human Feedback (RLHF). During the development of YaART, we especially focus on the choices of the model and training dataset sizes, the aspects that were not systematically investigated for text-to-image cascaded diffusion models before. In particular, we comprehensively analyze how these choices affect both the efficiency of the training process and the quality of the generated images, which are highly important in practice. Furthermore, we demonstrate that models trained on smaller datasets of higher-quality images can successfully compete with those trained on larger datasets, establishing a more efficient scenario of diffusion models training. From the quality perspective, YaART is consistently preferred by users over many existing state-of-the-art models.