Using reinforcement learning with human feedback (RLHF) has shown significant promise in fine-tuning diffusion models. Previous methods start by training a reward model that aligns with human preferences, then leverage RL techniques to fine-tune the underlying models. However, crafting an efficient reward model demands extensive datasets, optimal architecture, and manual hyperparameter tuning, making the process both time and cost-intensive. The direct preference optimization (DPO) method, effective in fine-tuning large language models, eliminates the necessity for a reward model. However, the extensive GPU memory requirement of the diffusion model's denoising process hinders the direct application of the DPO method. To address this issue, we introduce the Direct Preference for Denoising Diffusion Policy Optimization (D3PO) method to directly fine-tune diffusion models. The theoretical analysis demonstrates that although D3PO omits training a reward model, it effectively functions as the optimal reward model trained using human feedback data to guide the learning process. This approach requires no training of a reward model, proving to be more direct, cost-effective, and minimizing computational overhead. In experiments, our method uses the relative scale of objectives as a proxy for human preference, delivering comparable results to methods using ground-truth rewards. Moreover, D3PO demonstrates the ability to reduce image distortion rates and generate safer images, overcoming challenges lacking robust reward models.

翻译：利用强化学习与人类反馈（RLHF）在微调扩散模型方面展现出显著潜力。现有方法通常先训练一个符合人类偏好的奖励模型，再借助强化学习技术微调基础模型。然而，构建高效的奖励模型需要大规模数据集、最优架构及人工超参数调优，导致该过程耗时且成本高昂。尽管直接偏好优化（DPO）方法在微调大语言模型中效果显著，可免除奖励模型需求，但扩散模型去噪过程对GPU内存的极大需求阻碍了DPO方法的直接应用。为解决此问题，我们提出去噪扩散策略直接偏好优化（D3PO）方法，实现扩散模型的直接微调。理论分析表明，D3PO虽未显式训练奖励模型，却能等效于利用人类反馈数据训练的最优奖励模型来引导学习过程。该方法无需训练奖励模型，具有更直接、低成本且降低计算开销的优势。实验中，我们以目标的相对尺度作为人类偏好的代理指标，获得了与使用真实奖励的方法相当的结果。此外，D3PO展现出降低图像失真率、生成更安全图像的能力，有效克服了缺乏稳健奖励模型的挑战。