Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Mislabel detection on Weak Reference Labels Mislabels
Loading...
84.3
AP
4PL Δℓi (proposed)
52.06
60.43
68.8
77.17
May 28, 2026
AP
Precision@100
Precision@200
Updated 2d ago
Evaluation Results
Method
Method
Links
AP
Precision@100
Precision@200
4PL Δℓi (proposed)
Criterion=Forced-ceili...
2026.05
84.3
98
95
top-10 disagreement
2026.05
76.1
89
91.5
plain 4PL, single-stage (low di)
Model=4PL, Protocol=si...
2026.05
71.8
92.1
89.5
XGBoost (4PL params)
Model=XGBoost, Feature...
2026.05
71.7
93
90
4PL, low di
Model=4PL, Selection=l...
2026.05
70.3
95
87
low ri
Selection=low response...
2026.05
69.4
98
88
plain 2PL (low ai)
Model=2PL, Selection=l...
2026.05
64.2
95
72.5
GLAD
2026.05
62.2
94
81.5
overall disagreement
2026.05
53.3
80
73.5
Feedback
Search any
task
Search any
task