By Fabrice Popineau September 27, 2026
At the MultiTab workshop, held on September 27, 2026 in Strasbourg during MICCAI 2026 (September 27–October 1), our paper Single-Pass Conformal Cross-Modal Anomaly Screening with Tabular Foundation Models received the Best Paper Award.
José Lucas De Melo Costa, Seong Woo Ahn, Fabrice Popineau, Arpad Rimmel and Bich-Liên Doan use a pretrained tabular foundation model to detect records that look ordinary within each modality but deviate from the relationships between modalities. The method requires no additional training and improves false-positive calibration in the reported medical evaluations.
