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

Instruction Following on NoveltyBench

40.1Lexical Dominance

Aligned

-0.7729.83920.4531.061Nov 7, 2025
Updated 22h ago

Evaluation Results

MethodLinks
2025.11
40.127.312.817.22028.6
2025.11
3149.545.248.847.439.9
2025.11
9.814.214.213.114.211.4
2025.11
8--6.5-7.3
2025.11
6.819.216.17.617.67.2
2025.11
3.4--5.8-4.6
2025.11
0.8--1-0.9