List your company

Binary disease diagnosis on NIH-Google OOD

88.9Macro AccuracyCSRL
59.98867.4947582.506May 26, 2026

Evaluation results

MethodLinks
CSRL2026.0588.968.2
InternVL3.5-38B-Instruct2026.0587.947.8
Best Tool per Pathology2026.0586.667.4
Hulu-Med-Flash-Preview-27B2026.0585.466.6
Logistic Regression2026.0585.364.9
Gemini-3.1-Pro-Preview2026.0584.967
HuatuoGPT-Vision-7B2026.0584.753.5
GPT-4o2026.0584.550.9
Majority Voting (tie → neg)2026.058363.4
Qwen2.5-VL-7B-Instruct2026.0579.337.8
Qwen2.5-VL-72B-Instruct2026.0578.542.9
CheXzero2026.0578.161.4
MedGemma-27B-IT2026.0577.761.5
DenseNet-MIMIC2026.0577.449.1
Claude-Sonnet-4.62026.0575.452.3
CheXagent-8B2026.0574.661.4
Majority Voting (tie → pos)2026.0573.158.6
Lingshu-7B2026.0572.359.7
DenseNet-CheX2026.0570.252.2
MedKLIP2026.0567.756.5
DenseNet-NIH2026.0561.150