Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
RLHF on HH-RLHF (test)
Loading...
88.94
Average Score
GEOALIGN
41.2144
53.6047
65.995
78.3853
Jun 25, 2026
Average Score
Updated 29d ago
Evaluation Results
Method
Method
Links
Average Score
GEOALIGN
Backbone=Qwen3-4B
2026.06
88.94
GEOALIGN
Backbone=Qwen3-1.7B
2026.06
88.85
PF-PPO (BR)
Backbone=Qwen3-4B
2026.06
87.71
PAR
Backbone=Qwen3-4B
2026.06
87.61
SEED-GRPO
Backbone=Qwen3-4B
2026.06
87.43
BASE-GRPO
Backbone=Qwen3-4B
2026.06
86.72
PF-PPO (BW)
Backbone=Qwen3-4B
2026.06
86.6
PODS
Backbone=Qwen3-4B
2026.06
85.94
PF-PPO (BR)
Backbone=Qwen3-1.7B
2026.06
85.56
PAR
Backbone=Qwen3-1.7B
2026.06
84.91
SEED-GRPO
Backbone=Qwen3-1.7B
2026.06
84.24
PF-PPO (BW)
Backbone=Qwen3-1.7B
2026.06
83.96
BASE-GRPO
Backbone=Qwen3-1.7B
2026.06
83.54
PODS
Backbone=Qwen3-1.7B
2026.06
83.39
SFT
Backbone=Qwen3-4B
2026.06
44.73
SFT
Backbone=Qwen3-1.7B
2026.06
43.05
Feedback
Search any
task
Search any
task