Repulsive Latent Score Distillation for Solving Inverse Problems

Nicolas Zilberstein, Morteza Mardani, Santiago Segarra

Keywords: Inverse problems, Score distillation, Diffusion models, Mode collapse, Variational inference

TL;DR: We propose Repulsive Latent Score Distillation, a particle-based variational method to solve inverse problem at 512 x 512 resolution using latent diffusion models as prior.

Abstract:

Score Distillation Sampling (SDS) has been pivotal for leveraging pre-trained diffusion models in downstream tasks such as inverse problems, but it faces two major challenges: $(i)$ mode collapse and $(ii)$ latent space inversion, which become more pronounced in high-dimensional data.
To address mode collapse, we introduce a novel variational framework for posterior sampling.
Utilizing the Wasserstein gradient flow interpretation of SDS, we propose a multimodal variational approximation with a \emph{repulsion} mechanism that promotes diversity among particles by penalizing pairwise kernel-based similarity.
This repulsion acts as a simple regularizer, encouraging a more diverse set of solutions.
To mitigate latent space ambiguity, we extend this framework with an \emph{augmented} variational distribution that disentangles the latent and data.
This repulsive augmented formulation balances computational efficiency, quality, and diversity.
Extensive experiments on linear and nonlinear inverse tasks with high-resolution images ($512 \times 512$) using pre-trained Stable Diffusion models demonstrate the effectiveness of our approach.

Primary Area: generative models

Code Of Ethics: I acknowledge that I and all co-authors of this work have read and commit to adhering to the ICLR Code of Ethics.

Submission Guidelines: I certify that this submission complies with the submission instructions as described on https://iclr.cc/Conferences/2025/AuthorGuide.

Anonymous Url: I certify that there is no URL (e.g., github page) that could be used to find authors’ identity.

No Acknowledgement Section: I certify that there is no acknowledgement section in this submission for double blind review.

Submission Number: 5260