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

Instruction Following Evaluation on ArenaHard v1

38ArenaHardv1 Score

+RL (Skywork-Reward-V2-Llama-3.1-8B)

5.55213.97622.430.824Jul 2, 2025
Updated 1mo ago

Evaluation Results

MethodLinks
38
2025.07
37.9
2025.07
35
34.5
29.8
2025.07
24.9
2025.07
22.1
2025.07
20.8
2025.07
18.8
2025.07
16.2
2025.07
14
2025.07
12.6
2025.07
9.7
2025.07
6.8