List your company

Reward Modeling on Anthropic HH-helpful core250

0.292Delta RMTEA
0.192160.218080.2440.26992May 11, 2026

Evaluation results

MethodLinks
TEAN=64, Reward Model=Skywork/Skywork-Reward-V2-Qwen3-8B2026.050.29260.80.838.4
TEAN=32, Reward Model=Skywork/Skywork-Reward-V2-Qwen3-8B2026.050.24963.60.835.6
TEAN=128, Reward Model=Skywork/Skywork-Reward-V2-Qwen3-8B2026.050.24858.81.639.6
TEAN=16, Reward Model=Skywork/Skywork-Reward-V2-Qwen3-8B2026.050.21267.6032.4
TEAN=256, Reward Model=Skywork/Skywork-Reward-V2-Qwen3-8B2026.050.209563.240.8
TEAN=8, Reward Model=Skywork/Skywork-Reward-V2-Qwen3-8B2026.050.19669.6030.4