cs.CVNov 16, 2026

When Do Histology Foundation Models Need Geometric Test-Time Augmentation?

Authors: Lars Chen, Valentino F. Foit, Dorry Segev, Michal Mankowski

Organizations: Department of Surgery, New York University Langone Health, USA · Hertie Institute for AI in Brain Health, University of Tübingen, Germany · Department of Physics, New York University, USA

Abstract

Test-time augmentation (TTA) aggregates model predictions across multiple transformed views of an input during inference. Because histology images lack a canonical orientation, D4D_4 TTA---pooling across the 8 rotations and reflections of a square---is a standard default in computational pathology. However, the rise of self-supervised histology foundation models (FMs) with inherent rotational invariance calls its continued utility into question. We benchmark 9 histology FMs and 6 general-purpose models across four patch-level and two whole-slide image (WSI) datasets. We find that D4D_4 TTA remains a reliable post hoc improvement, but benefits diminish from +1.98+1.98 percentage points (pp) for general-purpose backbones to +0.50+0.50 pp for histology FMs on TCGA-UT, and shrink further for the largest models. To eliminate computational overhead, our experiments demonstrate that entropy routing---applying TTA selectively to uncertain predictions---recovers most of the gain while augmenting only ∼13%{\sim}13\% of samples. Finally, we show that patch-level TTA improvements do not transfer to slide-level multiple instance learning (MIL). Code: https://github.com/lars-chen/histology-tta

Figures & tables

Appendix figures & tables4 assets

Supplementary material from the paper’s appendix.

Appendix

Explore similar work

CardsList
  1. I Can't Believe TTA Is Not Better: When Test-Time Augmentation Hurts Medical Image Classification

    Apr 6, 2026Daniel Nobrega MedeirosTest-Time AugmentationData Augmentation

  2. Benchmarking Computational Pathology Foundation Models For Semantic Segmentation

    Feb 21, 2026Lavish Ramchandani, Aashay Tinaikar, Dev Kumar Das +2Histopathology Image SegmentationPathology Foundation Models

  3. Robustifying pathology foundation models via fine-tuning

    Jul 24, 2026Alexandre Filiot, Oskar Thaeter, Benoit Schmauch +1Computational PathologyFine-Tuning