System-Embedded Diffusion Bridge Models
This addresses inverse problems in science and engineering by improving supervised bridge methods, though it appears incremental as it builds on existing score-based generative models and bridge methods.
The paper tackles the problem of solving linear inverse problems by introducing System-Embedded Diffusion Bridge Models (SDBs), which embed known measurement systems into matrix-valued SDEs, resulting in consistent improvements across diverse applications and robust generalization under system misspecification.
Solving inverse problems -- recovering signals from incomplete or noisy measurements -- is fundamental in science and engineering. Score-based generative models (SGMs) have recently emerged as a powerful framework for this task. Two main paradigms have formed: unsupervised approaches that adapt pretrained generative models to inverse problems, and supervised bridge methods that train stochastic processes conditioned on paired clean and corrupted data. While the former typically assume knowledge of the measurement model, the latter have largely overlooked this structural information. We introduce System embedded Diffusion Bridge Models (SDBs), a new class of supervised bridge methods that explicitly embed the known linear measurement system into the coefficients of a matrix-valued SDE. This principled integration yields consistent improvements across diverse linear inverse problems and demonstrates robust generalization under system misspecification between training and deployment, offering a promising solution to real-world applications.