Diverse Sampling in Diffusion Models with Marginal Preserving Particle Guidance
Fuente:
arXiv
Saved in:
| Main Authors: | , , |
|---|---|
| Format: | Preprint |
| Published: |
2026
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866915989728264192 |
|---|---|
| author | Vinograd, Gal Achituve, Idan Fetaya, Ethan |
| author_facet | Vinograd, Gal Achituve, Idan Fetaya, Ethan |
| contents | 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. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2605_06553 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | Diverse Sampling in Diffusion Models with Marginal Preserving Particle Guidance Vinograd, Gal Achituve, Idan Fetaya, Ethan Machine Learning 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. |
| title | Diverse Sampling in Diffusion Models with Marginal Preserving Particle Guidance |
| topic | Machine Learning |
| url | https://arxiv.org/abs/2605.06553 |