arxivcs.LGnucl-exphysics.ins-det2026-06-28
Self-Supervised Calibration of Scientific Instruments Using Physical Consistency Constraints
Calibration remains one of the principal obstacles to the deployment of machine learning in scientific instrumentation because it typically relies on expert intervention, dedicated procedures, and manually labelled data. We introduce a physics-informed self-supervised framework t…