Noisy-label detection on MNIST benchmark leaderboard · SOTA2 Research

Noisy-label detection on MNIST

99.83Accuracy2D geometric metric

Evaluation Results

Method	Links
2D geometric metric2026.06		99.83	98.89	99.45
3D multi-metric2026.06		99.82	98.63	99.57
2D geometric metric2026.06		99.77	99.44	99.4
3D multi-metric2026.06		99.77	99.42	99.42
2D geometric metric2026.06		99.69	99.58	99.64
3D multi-metric2026.06		99.69	99.58	99.65
Cleanlab2026.06		99.62	95.52	96.86
Cleanlab2026.06		99.57	97.66	98.07
2D geometric metric2026.06		99.49	97.77	99.72
Cleanlab2026.06		99.35	98.57	98.18
2D geometric metric2026.06		99.28	98.52	99.7
Cleanlab2026.06		99.1	99.13	98.62
2D geometric metric2026.06		96.55	59.16	99.83
3D multi-metric2026.06		96.55	92.07	99.97
3D multi-metric2026.06		93.2	42.34	99.97
2D geometric metric2026.06		92.7	40.61	99.97
2D geometric metric2026.06		92.42	56.86	99.98
3D multi-metric2026.06		90.2	67.12	100
3D multi-metric2026.06		89.32	31.85	99.97
3D multi-metric2026.06		88.91	47.39	100