List your company

Question Answering on PubMedQA (AR%, Risk%)

100AR (%)Fixed-Threshold
-2.33624.23250.877.368May 18, 2026

Evaluation results

MethodLinks
Fixed-Thresholdalpha=0.202026.0510023.9
Always-Actalpha=0.202026.0510023.9
ACIalpha=0.202026.0584.921.8
NEX-Confalpha=0.202026.0581.918.8
SAOCPalpha=0.202026.0581.222.4
CSAalpha=0.202026.0563.515.1
ConfFactalpha=0.202026.0534.87.6
Naive-Tuningalpha=0.202026.051.624.9