Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Alignment on HH IDN 40%
Loading...
68
Win Rate
UDPO
45.12
51.06
57
62.94
Jul 3, 2026
Win Rate
Updated 18d ago
Evaluation Results
Method
Method
Links
Win Rate
UDPO
Base Model=Qwen-3-1.7B
2026.07
68
α-UDPO
Base Model=Qwen-3-1.7B
2026.07
67.6
rDPO
Base Model=Qwen-3-1.7B
2026.07
63.7
DPO
Base Model=Qwen-3-1.7B
2026.07
56.4
α-URM
Base Model=Qwen-3-1.7B
2026.07
55.2
URM
Base Model=Qwen-3-1.7B
2026.07
54.6
rRM
Base Model=Qwen-3-1.7B
2026.07
48.2
RM
Base Model=Qwen-3-1.7B
2026.07
46
Feedback
Search any
task
Search any
task