Shortest-Path Flow Matching with Mixture-Conditioned Bases for OOD Generalization to Unseen Conditions
This addresses the problem of out-of-distribution generalization for researchers and practitioners in conditional generative modeling, offering a plug-in strategy that is incremental in nature.
The paper tackled the challenge of robust generalization under distribution shift in conditional generative modeling by introducing SP-FM, a shortest-path flow-matching framework that conditions both the base distribution and flow field on the condition, enabling smooth interpolation and reliable extrapolation to unseen conditions. Empirically, SP-FM was effective in domains such as single-cell transcriptomics and drug screening, though no concrete numerical results were provided in the abstract.
Robust generalization under distribution shift remains a key challenge for conditional generative modeling: conditional flow-based methods often fit the training conditions well but fail to extrapolate to unseen ones. We introduce SP-FM, a shortest-path flow-matching framework that improves out-of-distribution (OOD) generalization by conditioning both the base distribution and the flow field on the condition. Specifically, SP-FM learns a condition-dependent base distribution parameterized as a flexible, learnable mixture, together with a condition-dependent vector field trained via shortest-path flow matching. Conditioning the base allows the model to adapt its starting distribution across conditions, enabling smooth interpolation and more reliable extrapolation beyond the observed training range. We provide theoretical insights into the resulting conditional transport and show how mixture-conditioned bases enhance robustness under shift. Empirically, SP-FM is effective across heterogeneous domains, including predicting responses to unseen perturbations in single-cell transcriptomics and modeling treatment effects in high-content microscopy--based drug screening. Overall, SP-FM provides a simple yet effective plug-in strategy for improving conditional generative modeling and OOD generalization across diverse domains.