Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Prototypicality Bias Evaluation on ProtoBias Objects
Loading...
0.653
Correct Ranking Margin
GPT-5
0.16212
0.28956
0.417
0.54444
Jan 8, 2026
Correct Ranking Margin
Incorrect Ranking Margin
Updated 1mo ago
Evaluation Results
Method
Method
Links
Correct Ranking Margin
Incorrect Ranking Margin
GPT-5
2026.01
0.653
0
GPT-4o
2026.01
0.565
0.167
PROTOSCORE
2026.01
0.346
0.038
VQAScore
2026.01
0.225
0.018
CLIPScore
2026.01
0.196
0.096
PickScore
2026.01
0.181
0.09
Feedback
Search any
task
Search any
task
