Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Preference Learning on Toy dataset Noise 10% (test)
Loading...
93.1
Accuracy
SSPO
58.156
67.228
76.3
85.372
Oct 28, 2025
Accuracy
Updated 1mo ago
Evaluation Results
Method
Method
Links
Accuracy
SSPO
prior=0.5, n_L=100, n_...
2025.10
93.1
SSPO
prior=0.5, n_L=10, n_U...
2025.10
84
SSPO
prior=0.5, n_L=50, n_U...
2025.10
81.2
DPO
n_L=100, n_U=1000
2025.10
78.3
SimPO
n_L=100, n_U=1000
2025.10
77
SimPO
n_L=10, n_U=1000
2025.10
74.4
DPO
n_L=50, n_U=1000
2025.10
74.2
SimPO
n_L=50, n_U=1000
2025.10
73.7
DPO
n_L=10, n_U=1000
2025.10
69.5
ORPO
n_L=100, n_U=1000
2025.10
68.2
ORPO
n_L=50, n_U=1000
2025.10
64.6
ORPO
n_L=10, n_U=1000
2025.10
59.5
Feedback
Search any
task
Search any
task
