# Physics-Grounded Adversarial Stain Augmentation with Calibrated Coverage Guarantees

arXiv: https://arxiv.org/abs/2605.13889 · HTML: https://mlsanity.com/paper/2605.13889/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 6 of its 10 references that are in the map.
- First posted on arXiv 12 May 2026.

## Builds on (most cited first, 6 of 6)

- [Domain-Adversarial Training of Neural Networks](https://mlsanity.com/paper/1505.07818/) (2015)
- [WILDS: A Benchmark of in-the-Wild Distribution Shifts](https://mlsanity.com/paper/2012.07421/) (2020)
- [Quantifying the effects of data augmentation and stain color normalization in convolutional neural networks for computational pathology](https://mlsanity.com/paper/1902.06543/) (2019)
- [Adversarial Style Augmentation for Domain Generalized Urban-Scene Segmentation](https://mlsanity.com/paper/2207.04892/) (2022)
- [RandStainNA: Learning Stain-Agnostic Features from Histology Slides by Bridging Stain Augmentation and Normalization](https://mlsanity.com/paper/2206.12694/) (2022)
- [AdvST: Revisiting Data Augmentations for Single Domain Generalization](https://mlsanity.com/paper/2312.12720/) (2023)

## Abstract (arXiv)

Stain variation across hospitals degrades histopathology models at deployment. Existing augmentation methods perturb color spaces with arbitrary hyperparameters, lacking both a principled budget and coverage guarantees for unseen centers. We propose Calibrated Adversarial Stain Augmentation (CASA), which performs adversarial augmentation in the Macenko stain parameter space with a budget calibrated from multi-center statistics via the DKW inequality. On Camelyon17-WILDS (5 seeds), CASA achieves 93.9% ± 1.6% slide-level accuracy -- outperforming HED-strong (88.4% ± 7.3%), RandStainNA (85.2% ± 6.7%), and ERM (63.9% ± 11.3%) -- with the highest worst-group accuracy (84.9% ± 0.9%) among all 10 compared methods.