Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Preference Discrimination on PREFBENCH
Loading...
98.67
Top-1 Acc
Human
80.938
85.5415
90.145
94.7485
Dec 4, 2025
Top-1 Acc
Updated 1mo ago
Evaluation Results
Method
Method
Links
Top-1 Acc
Human
type=experts
2025.12
98.67
PrefDisc
2025.12
98.67
Claude-3.5-Sonnet
2025.12
90.44
ViPer Proxy Metric
2025.12
81.62
Feedback
Search any
task
Search any
task