SEIKO is a novel reinforcement learning method to efficiently fine-tune diffusion models in an online setting. Our methods outperform all baselines (PPO, classifier-based guidance, direct reward backpropagation) for fine-tuning Stable Diffusion.
Hi, Zhao Yulai, I noticed that in another article Bridging Model-Based Optimization and Generative Modeling via Conservative Fine-Tuning of Diffusion Models, there seems to be no mention of the relevant code base. Can you also share the code of this article? Thank you very much.
Hi, Zhao Yulai, I noticed that in another article Bridging Model-Based Optimization and Generative Modeling via Conservative Fine-Tuning of Diffusion Models, there seems to be no mention of the relevant code base. Can you also share the code of this article? Thank you very much.