Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Reward Model Evaluation on Alpaca-Eval
Loading...
72.59
LC Win Rate
FiMi-RM
62.294
64.967
67.64
70.313
May 19, 2025
LC Win Rate
Win Rate
Length (Characters)
Length (Tokens)
Updated 1mo ago
Evaluation Results
Method
Method
Links
LC Win Rate
Win Rate
Length (Characters)
Length (Tokens)
FiMi-RM
Backbone=Qwen2.5-7B, A...
2025.05
72.59
76.39
543
125
ODIN
Backbone=Qwen2.5-7B, A...
2025.05
71.38
75.58
512
115
Length Penalty
Backbone=Qwen2.5-7B, A...
2025.05
69.22
74.54
620
144
Vanilla RM
Backbone=Qwen2.5-7B, A...
2025.05
68.25
73.84
638
148
FiMi-RM
Backbone=Gemma2-9B, Al...
2025.05
66.68
67.77
534
123
ODIN
Backbone=Gemma2-9B, Al...
2025.05
63.46
65.24
684
168
Vanilla RM
Backbone=Gemma2-9B, Al...
2025.05
62.91
65.77
756
179
Length Penalty
Backbone=Gemma2-9B, Al...
2025.05
62.69
65.62
712
158
Feedback
Search any
task
Search any
task