A benchmark study of vision and pathology foundation models for computational pathology
Rohan Bareja, Francisco Carrillo-Perez, Yuanning Zheng, Marija Pizurica, Tarak Nath Nandi, Lu Tian, Jeanne Shen, Ravi Madduri, Olivier Gevaert · 2026
Read A benchmark study of vision and pathology foundation models for computational pathology by Rohan Bareja, Francisco Carrillo-Perez, Yuanning Zheng, Marija Pizurica, Tarak Nath Nandi, Lu Tian, Jeanne Shen, Ravi Madduri, Olivier Gevaert (2026) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/PMC13507134.txt.
Description
Rohan Bareja and colleagues benchmark, in 2026 work, thirty-two AI foundation models for computational pathology, the use of computers to read tissue slides. Models spanned general vision, general vision-language, and pathology-specific versions of both types. Tests used cancer slide tasks at whole-slide and small-patch levels from major public cohorts and outside datasets.
Pathology vision models ranked among the strongest on the main cohort, and beat pathology vision-language models overall. Rankings shifted modestly across outside datasets, and gaps among top models were often small. Bigger models and larger training sets did not reliably win. Blending model votes late in the pipeline lifted scores across sites and tissues. No single model fits every task.
Catalog record
| CREATOR | Rohan Bareja, Francisco Carrillo-Perez, Yuanning Zheng, Marija Pizurica, Tarak Nath Nandi, Lu Tian, Jeanne Shen, Ravi Madduri, Olivier Gevaert |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 119 KB |
| LICENSE | https://creativecommons.org/licenses/by/4.0 |
| RIGHTS | CC BY |
| SOURCE | Europe PMC |