COMPASS Beats 22 Immunotherapy Prediction Methods in Leave-One-Out Evaluation
marinkazitnik · x · 2026-07-03
In leave-one-out cross-validation, COMPASS improved average accuracy by 8.5% and precision-recall AUC by 15.7%, outperforming 22 commonly used biomarkers and predictive models while achieving the highest cross-cohort transfer generalization rate. Even when completely excluding a specific cancer type, drug class, or checkpoint target during training, the model maintained its performance on the excluded categories, indicating its ability to capture anti-tumor immune biological patterns transferable across cancer types.
More from Research
- MissionReport benchmark: only 3 of 540 AI-generated government reports pass all criteria — Exp_Mark · 2026-10-27
- Finch 1: a geospatial foundation model trained on imagery from 50+ sensors — Scobleizer · 2026-10-27
- Skyfall AI's SKY-001 world model turns sketches into CAD builds in 3.9s, vs 199s for GPT-6 Astra — rohanpaul_ai · 2026-10-27
- Physicists Let AI Pick Its Own Problem—It Wrote a Publishable Paper in 2 Hours — burny_tech · 2026-10-27
- Researcher flags unresolved data transparency issues in compositionality evals — najoungkim · 2026-10-27
- Separate Forms from Estimators: Why Steer by Difference of Means When You Can Use SGD? — johnhewtt · 2026-10-27