Blind Strong Gravitational Lensing Inversion: Joint Inference of Source and Lens Mass with Score-Based Models

TL;DR

This paper introduces a novel method for jointly inferring both the source galaxy and the lens mass distribution in strong gravitational lensing using score-based models, enabling more accurate and unbiased reconstructions.

Contribution

It is the first to demonstrate joint source and lens inference in gravitational lensing with a score-based prior, relaxing previous assumptions of known lens mass.

Findings

Reconstructed sources match observational noise levels.

Marginal posteriors of lens parameters recover true values without bias.

First successful joint inference of source and lens in this context.

Abstract

Score-based models can serve as expressive, data-driven priors for scientific inverse problems. In strong gravitational lensing, they enable posterior inference of a background galaxy from its distorted, multiply-imaged observation. Previous work, however, assumes that the lens mass distribution (and thus the forward operator) is known. We relax this assumption by jointly inferring the source and a parametric lens-mass profile, using a sampler based on GibbsDDRM but operating in continuous time. The resulting reconstructions yield residuals consistent with the observational noise, and the marginal posteriors of the lens parameters recover true values without systematic bias. To our knowledge, this is the first successful demonstration of joint source-and-lens inference with a score-based prior.