Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Multi-objective Alignment on Human Values (full evaluation set)
Loading...
67.64
Help Score
MOPPO
57.6144
60.2172
62.82
65.4228
Oct 1, 2025
Help Score
Honesty Score
Truthfulness Score
Updated 1d ago
Evaluation Results
Method
Method
Links
Help Score
Honesty Score
Truthfulness Score
MOPPO
Objective=Scalarized w...
2025.10
67.64
29.21
14.6
MAH-DPO Ensemble
Mode=Ensemble
2025.10
63.89
36.87
24.78
MODPO
2025.10
61.75
34.77
23.25
DPO Soup
2025.10
61.28
32.17
21.53
Base
2025.10
58
30.42
18.88
Feedback
Search any
task
Search any
task