用温度采样提升扩散模型多样性，解决样本质量下降问题。
Diversify Diffusion with Temperature Sampling and Variance-Corrective Time Shifting
- 通过温度采样与时间偏移校正，无须重训练即可调节生成多样性。
- 在多种模型上实现多样性提升，且对图像质量和条件一致性影响极小。
- 支持从粗到细的控制：高噪声阶段生成组合多样性，低噪声阶段细化细节。
扩散模型忠实再现训练分布，但也继承其不平衡性，使罕见模式难以生成。推理阶段采用高温目标分布 $p^{(γ)}_0(x) /propto p_0(x)^γ$（$0 < γ < 1$）可平坦主导模式并提升稀有模式概率。但直接缩放得分会放大各模式方差，破坏反向扩散过程，导致样本质量下降。本文提出无训练的方差校正时间偏移方法：在偏移的时间步查询网络，并将得分缩放 $γ$，抵消方差膨胀，同时保留模式重加权效果。该修正使简单温度采样成为预训练扩散模型和流匹配骨干网络的实用多样性调控工具，且在 DiT、Stable Diffusion 与 Motion Diffusion 模型中均实现一致增益，代价极小。此外，温度干预时机可实现粗到细控制：高噪声阶段促进跨模式组合多样性，低噪声阶段在固定组合下生成局部外观差异。
原文摘要 · Abstract (English)
Diffusion models faithfully reproduce their training distribution, but also inherit its imbalances and leave rare or under-represented modes hard to reach. A natural inference-time remedy is to sample from the high-temperature target $p^{(γ)}_0(x) \propto p_0(x)^γ$ for $0 < γ< 1$, which flattens dominant modes and lifts rare ones. However, naive score scaling while correctly reweighting modes also inflates the per-mode variance, breaking the reverse diffusion process and degrading sample quality. We introduce variance-corrective time shifting, a training-free fix that queries the network at a shifted timestep and scales the resulting score by $γ$, canceling the variance inflation while preserving the mode reweighting. The correction turns simple temperature sampling into a practical diversity knob for pretrained diffusion and flow-matching backbones with no retraining, and we demonstrate consistent gains at minimal cost to sample quality and condition fidelity across DiT, Stable Diffusion and Motion Diffusion models. We further show that the timing of the temperature intervention enables coarse-to-fine control: high-noise stages drive compositional diversity across modes, while low-noise stages drive local appearance variation under a fixed composition.
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。