arXiv:2607. 25497v1 Announce Type: cross Abstract: Pathology foundation models are approaching clinical deployment, yet remain vulnerable to systematic non-biological variation across centres.
arXiv:2607. 25497v2 Announce Type: replace-cross Abstract: Pathology foundation models encode non-biological variation introduced by tissue preparation, staining and scanning, enabling shortcut learning that undermines generalisation across institutions.
HERO (Histology Encoder for Robust Representation in Oncology) is a ViT‑G/14 pathology foundation model trained with DINO and iBOT objectives and refined using high‑resolution Gram anchoring on a 500‑million‑tile corpus from about 575,000 clinical whole‑slide images. It demonstrates superior robustness to center, scanner, and stain variation compared to other state‑of‑the‑art foundation models, while maintaining competitive performance on tile‑level classification, segmentation, and gene‑expression prediction. Across 39 slide‑level clinical tasks, HERO ranks first on average and achieves the best average rank across six benchmark frameworks under an equal‑weighted analysis.
arXiv:2607. 04401v1 Announce Type: cross Abstract: How robust and generalisable are pathology foundation models and have their scaling limites been reached?
arXiv:2606. 07590v1 Announce Type: cross Abstract: Pathology foundation models are pretrained on large streams of WSI-derived patches, while supervision during data construction is often slide-level, sparse, or heterogeneous.
The study evaluates whether disease can be identified from reactive, non‑lesional brain tissue in intracranial biopsies. Using four foundation‑model encoders within an attention‑based multiple‑instance learning framework on 245 whole‑slide images, the authors find that disease labels remain predictive even after controlling for slide size and sampling bias, and that performance is similar across all encoders. Signed instance‑contribution maps and expert review confirm that predictive signals localize to reactive parenchyma rather than artifacts such as blood. "whyItMatters":"The findings demonstrate that weakly supervised models can recover disease signals from tissue traditionally considered non‑diagnostic, highlighting the need for provenance‑only baselines in computational pathology benchmarks."
arXiv:2609.32876v2 Announce Type: replace-cross Abstract: State-of-the-art pathology foundation models, trained on millions of histology tiles, can fail to preserve tissue similarity when comparisons...
The study investigates whether pathology foundation models (PFMs) carry center-related biases into whole-slide image (WSI) classification. By training models with increasing class-center correlations and evaluating six PFMs across four datasets and two MIL aggregators, the authors introduce the Area Under the Cramér's V Curve (AUCC) to measure both accuracy and degradation due to spurious correlations. Results reveal that center information propagates to WSI predictions, with robustness varying by PFM and MIL strategy, and that ComBat harmonization does not consistently improve robustness.
arXiv:2505. 16941v4 Announce Type: replace-cross Abstract: Foundation models (FMs) promise to address core limitations of traditional supervised machine learning: (i) reliance on large amounts of labeled data, (ii) task specificity, and (iii) poor transportability.
arXiv:2608. 05960v1 Announce Type: cross Abstract: Routine CT interpretation is inherently comprehensive, capturing incidental findings across the entire scan volume.
The study evaluates whether disease can be identified from reactive, non‑lesional brain tissue in intracranial biopsies. Using four foundation‑model encoders as frozen patch encoders within an attention‑based multiple‑instance learning framework, the authors benchmark performance on 245 whole‑slide images from 186 patients. They find that coarse disease‑category prediction can be largely explained by slide size, but finer diagnostic distinctions remain predictive above chance, with no significant difference among the encoders, suggesting weak morphological signals can be recovered even from tissue traditionally considered non‑diagnostic.
CellPath-Bench is a new benchmark that evaluates whole-slide cellular representations in pathology foundation models (PFMs) by using 25 spatially aligned H&E–Xenium tissue sections from 11 organs and over 7 million cells. It introduces metrics such as Cell Representation Advantage (CRA) and Cell Representation Transferability (CRT) to assess how well frozen PFMs encode cell-type information and generalize across tissue sections, datasets, and organs. The benchmark was applied to 30 PFMs, revealing significant model-dependent differences in cell-type decodability and cross-domain generalization, and offers a standardized framework for auditing cellular information in frozen PFM representations.
arXiv:2505.03380v2 Announce Type: replace Abstract: Accurate delineation of tumors and surrounding organs-at-risk is essential for radiotherapy, surgery and treatment response assessment, yet remains...