Share your thoughts, 1 month free Claude Pro on usSee more

Scientific Question Answering on MMLU

79.45Accuracy

EVOSELECT

Updated 1mo ago

Evaluation Results

Method	Links
EVOSELECT 2026.04		79.45
Attr-Div 2026.04		79.28
Random 2026.04		79.22
EVOSELECT 2026.04		79.22
Attribution 2026.04		79.19
EVOSELECT 2026.04		79.18
Random 2026.04		79.18
Diversity 2026.04		79.15
EVOSELECT 2026.04		79.14
TSDS 2026.04		79.13
TSDS 2026.04		79.12
All 2026.04		79.11
Attribution 2026.04		79.07
Attribution 2026.04		79.06
TSDS 2026.04		79.03
Diversity 2026.04		79.03
Attr-Div 2026.04		78.91
Attribution 2026.04		78.89
Diversity 2026.04		78.86
Attr-Div 2026.04		78.86
Base 2026.04		78.82
Random 2026.04		78.79
Diversity 2026.04		78.75
TSDS 2026.04		78.75
Random 2026.04		78.66
All 2026.04		78.52
Attr-Div 2026.04		78.52