让扩散模型生成更多样化的图像，同时保持质量不下降。
Diverse Sampling in Diffusion Models with Marginal Preserving Particle Guidance
- 通过保持边缘分布不变的粒子引导机制，提升生成多样性。
- 在文本到图像生成中，多样性提升显著，且分布保真度高。
- 无需额外训练，适用于高维生成任务，计算高效。
我们提出EDDY（Exact-marginal Diversification via Divergence-free dYnamics），一种用于扩散模型和流匹配模型的引导机制，能在保持生成样本质量的同时提升多样性。EDDY利用福克-普朗克方程的对称性，采用改变粒子轨迹但保持边缘分布不变的漂移扰动。通过基于核函数的反称成对矩阵场实现，该方法构建于排斥方向之上，产生无散度动力学，在联合粒子层面促进多样性，同时不改变单个粒子的边际分布，且无需额外训练。针对如文本到图像生成中使用感知嵌入时计算成本高的问题，我们提出了实用的近似方案，实现高效有效。在合成分布与文本到图像生成上的实验表明，相比常见基线，EDDY在保持强分布保真度的同时显著提升多样性。
原文摘要 · Abstract (English)
We present EDDY (Exact-marginal Diversification via Divergence-free dYnamics), a guidance mechanism for diffusion and flow matching models that promotes diversity among samples generated while maintaining quality. EDDY exploits symmetries of the Fokker-Planck equation, using drift perturbations that change particle trajectories while preserving the evolving marginal distribution. We instantiate this principle through kernel-based anti-symmetric pairwise matrix fields, constructed from the repulsive directions. The resulting divergence-free dynamics promote diversity at the joint particle level while preserving each particle's marginal distribution without any additional training. As computing the guidance can be computationally expensive in cases such as text-to-image generation with perceptual embeddings, we propose practical approximations as an effective and efficient solution. Experiments on synthetic distributions and text-to-image generation show that EDDY improves diversity while maintaining strong distributional fidelity compared to common baselines.
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。