新方法让扩散模型更好找数据模式，生成更真实图像。
Mean-Shift Distillation for Diffusion Mode Seeking
- 用均值漂移思想直接逼近扩散模型输出分布的梯度
- 在文本到图像和3D生成中模式对齐更准，收敛更快
- 可直接替换现有采样方法，无需重训练
我们提出均值漂移蒸馏，一种新颖的扩散蒸馏技术，能提供扩散输出分布梯度的可证明良好代理。该方法直接基于分布上的均值漂移模式搜索推导而来，且其极值与模式对齐。我们进一步推导出一种高效的乘积分布采样过程以评估梯度。本方法可作为分数蒸馏采样（SDS）的即插即用替代方案，无需模型重训练或采样流程大幅修改。实验表明，该方法在合成与实际场景中均表现出更优的模式对齐效果和更快的收敛速度，在Stable Diffusion的文本到图像及文本到3D应用中生成了更高保真度的结果。
原文摘要 · Abstract (English)
We present mean-shift distillation, a novel diffusion distillation technique that provides a provably good proxy for the gradient of the diffusion output distribution. This is derived directly from mean-shift mode seeking on the distribution, and we show that its extrema are aligned with the modes. We further derive an efficient product distribution sampling procedure to evaluate the gradient. Our method is formulated as a drop-in replacement for score distillation sampling (SDS), requiring neither model retraining nor extensive modification of the sampling procedure. We show that it exhibits superior mode alignment as well as improved convergence in both synthetic and practical setups, yielding higher-fidelity results when applied to both text-to-image and text-to-3D applications with Stable Diffusion.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。