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

Multi-turn dialogue on MT-bench

5.25Kendall's Tau

CORAL

2.132.943.754.56Feb 24, 2025
Updated 1mo ago

Evaluation Results

MethodLinks
2025.02
5.253.45
2025.02
5.23.42
2025.02
5.13.32
2025.02
5.093.13
2025.02
5.023.26
2025.02
5.023.09
2025.02
4.862.94
2025.02
4.83.16
2025.02
4.772.88
2025.02
4.683.06
2025.02
4.682.89
2025.02
4.572.5
2025.02
4.562.62
2025.02
4.522.59
2025.02
4.482.45
2025.02
4.452.7
2025.02
4.262.63
2025.02
4.262.74
2025.02
4.232.79
2025.02
4.222.3
2025.02
4.162.28
2025.02
4.152.24
2025.02
4.152.26
2025.02
4.132.72
2025.02
4.132.35
2025.02
4.122.21
2025.02
4.122.35
2025.02
4.12.71
2025.02
4.082.7
2025.02
4.082.36
2025.02
4.022.5
2025.02
42.65
2025.02
3.933.04
2025.02
3.922.39
2025.02
3.912.61
2025.02
3.912.13
2025.02
3.882.46
2025.02
3.842.08
2025.02
3.832.05
2025.02
3.82.67
2025.02
3.762.15
2025.02
3.651.96
2025.02
3.642.65
2025.02
3.622.37
2025.02
3.622.21
2025.02
3.451.86
2025.02
2.872.24
2025.02
2.672.06
2025.02
2.631.65
2025.02
2.631.83
2025.02
2.531.56
2025.02
2.441.62
2025.02
2.41.41
2025.02
2.251.27