Published September 24, 2026
| Version v3.0 (GitHub snapshot, code only)
Software
Open
cpath-triage: diagnostic analysis of verbalized confidence in a medical vision-language model
Authors/Creators
- 1. Department of Computer and Information Science and Engineering, University of Florida, Gainesville, FL, U.S.A.
- 2. Department of Computer Science and Engineering, JK Lakshmipat University, Jaipur, India
- 3. Department of Data Science and Engineering, Manipal University Jaipur, Jaipur, India
- 4. Department of Computer Applications, Manipal University Jaipur, Jaipur, India
Description
GitHub snapshot, code only. This record was created automatically from the GitHub tag v3.0 and holds the source code without the prediction artifacts, so it cannot regenerate the paper's tables on its own. The full archives are v3.0, doi:10.5281/zenodo.22927577, and the current release v4.0, doi:10.5281/zenodo.23199902.
Evaluation and analysis code, plus the prediction artifacts behind every table and figure, for a diagnostic study of verbalized confidence in MedGemma-27b-it on colorectal H&E histopathology patches. Covers zero-shot classification, temperature scaling and calibration metrics, selective-prediction routing with explicit tie handling, a design-effect sensitivity analysis in place of clustered intervals, and K=5 prompt-variant consistency routing.Notes
Code only. For the prediction artifacts, use the full archive, doi:10.5281/zenodo.23199902.
Files
anayy09/cpath-triage-v3.0.zip
Additional details
Related works
- Is supplement to
- Software: https://github.com/anayy09/cpath-triage/tree/v3.0 (URL)
Software
- Repository URL
- https://github.com/anayy09/cpath-triage