Clinical EvalEvidence / original sources
Follow the references behind Clinical Eval. Each entry explains its role, links to the original publication, and identifies the guides that use it.
Content updated September 28, 2026. Inclusion does not imply endorsement.
Chaoyi Wu and colleagues · npj Digital Medicine
Final journal version: benchmark composition, evaluation settings, task-specific metrics and results.
Version: Published 2025-01-27
Evidence locator: Figure 1; Results; Tables 6–7
www.nature.comMAGIC / Henrychur
Official task taxonomy and distinction between full source data and the reproduction split.
Version: Accessed 2026-09-28
Evidence locator: Introduction; sampling note
huggingface.coMAGIC: Medical Artificial General Intelligence Consortium
Author code; distinguishes training instances from benchmark data. README retains an earlier source-dataset count.
Version: Commit 185cfc04a27023983e48b357a6f2fd525049b9af
Evidence locator: README; LICENSE
github.comMAGIC / Henrychur
Published selected examples used for reproduction; do not silently substitute fresh random samples.
Version: Accessed 2026-09-28
Evidence locator: Files and versions
huggingface.coMAGIC
Current public scorer normalizes text and checks reference containment in the response.
Version: Pinned 2025-05-05 commit
Evidence locator: information_extraction_accuracy; evaluate_information_extraction
github.comMAGIC
Current public script aggregates entity TP, FP and FN within tasks and averages task F1 values.
Version: Pinned 2025-05-05 commit
Evidence locator: evaluate_hard_ner
github.com