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

Human Consistency Evaluation on RichHF-18K

33.9Kendall's Tau

Gemini-2.5-Pro

1.9210.222518.52526.8275Jun 3, 2025
Updated 1mo ago

Evaluation Results

MethodLinks
2025.06
33.9
2025.06
33.7
2025.06
33.6
2025.06
33.4
2025.06
31.5
2025.06
28.9
2025.06
27.1
2025.06
24.3
2025.06
16.8
2025.06
6.55
2025.06
3.15