New at AISTATS: IUPM — a label-free method for reliable model monitoring under drift.

We propose IUPM, a label-free method for tracking performance under gradual distribution shifts using optimal transport. IUPM quantifies uncertainty in its estimates and guides targeted labeling to restore reliability, outperforming existing baselines across scenarios. Published in the Proceedings of the 28th International Conference on Artificial Intelligence and Statistics (AISTATS).