AI RESEARCH
Uncertainty Quantification for Distribution-to-Distribution Flow Matching in Scientific Imaging
arXiv CS.LG
•
ArXi:2603.21717v1 Announce Type: new Distribution-to-distribution generative models scientific imaging tasks ranging from modeling cellular perturbation responses to translating medical images across conditions. Trustworthy generation requires both reliability (generalization across labs, devices, and experimental conditions) and accountability (detecting out-of-distribution cases where predictions may be unreliable