Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Model Utility Preservation on Model Utility (retain set)
Loading...
61
Utility
NPO
53.72
55.61
57.5
59.39
Oct 20, 2025
Utility
Updated 6d ago
Evaluation Results
Method
Method
Links
Utility
NPO
Model=Qwen3-8B, Varian...
2025.10
61
UNDIAL
Model=Qwen3-8B, Varian...
2025.10
61
RMU
Model=Gemma-2B-IT, Var...
2025.10
60
NPO
Model=Qwen3-8B, Varian...
2025.10
60
UNDIAL
Model=Qwen3-8B, Varian...
2025.10
60
RMU
Model=Qwen3-8B, Varian...
2025.10
59
NPO
Model=Gemma-2B-IT, Var...
2025.10
57
NPO
Model=Gemma-2B-IT, Var...
2025.10
57
RMU
Model=Gemma-2B-IT, Var...
2025.10
57
RMU
Model=Qwen3-8B, Varian...
2025.10
57
UNDIAL
Model=Gemma-2B-IT, Var...
2025.10
55
UNDIAL
Model=Gemma-2B-IT, Var...
2025.10
54
Feedback
Search any
task
Search any
task