Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Preference Labeling on Anthropic Helpfulness
Loading...
81
Preference Labeling Accuracy
Curriculum-RLAIF
57.08
63.29
69.5
75.71
May 26, 2025
Preference Labeling Accuracy
Updated 1mo ago
Evaluation Results
Method
Method
Links
Preference Labeling Accuracy
Curriculum-RLAIF
Base Model=LLaMA-3-8B
2025.05
81
RLCD
Base Model=LLaMA-3-8B
2025.05
77
Conventional RLAIF
Base Model=LLaMA-3-8B
2025.05
76
Curriculum-RLAIF
Base Model=Gemma-1-2B
2025.05
72
Conventional RLAIF
Base Model=Gemma-1-2B
2025.05
69
RLCD
Base Model=Gemma-1-2B
2025.05
67
CAI
Base Model=LLaMA-3-8B
2025.05
62
CAI
Base Model=Gemma-1-2B
2025.05
58
Feedback
Search any
task
Search any
task