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

Multi-Category Quality Scoring on MT-Bench

8.03Overall Score

AdaDPO

3.02764.32635.6256.9237Nov 15, 2024Feb 16, 2025May 20, 2025Aug 21, 2025Nov 22, 2025Feb 23, 2026May 27, 2026
Updated 6d ago

Evaluation Results

MethodLinks
2026.05
8.03--------
2026.05
8.01--------
2026.05
7.58--------
2024.11
6.389.257.24.652.553.35.558.939.58
2024.11
5.738.25.984.52.53.055.167.78.7
2024.11
5.257.27.033.951.71.74.357.538.53
2024.11
3.224.34.052.351.91.251.555.235.15