Image Classification on CIFAR-10 (test) (Thresholded Accuracy Evaluation) benchmark leaderboard · SOTA2 Research

Image Classification on CIFAR-10 (test) (Thresholded Accuracy Evaluation)

92.48Test Accuracy (Symmetric, η=0.2)Robust Self-Training with Closed-loop Label Correction

Evaluation Results

Method
Robust Self-Training with Closed-loop Label Correction2026.03	92.48	89.28	85.5	93.88	92.65	93.07	92.83	91.37	85.76
Robust Self-Training with Closed-loop Label Correction2026.03	92.12	88.73	84.61	93.7	92.61	92.97	92.69	91.09	85.24
FasTEN2026.03	91.46	88.42	82.88	92.87	92.68	91.39	91.19	85.44	77.09
Cross-Entropy2026.03	91.13	87.93	82.54	92.87	90	71.99	90.75	82.74	68.52
EMLC2026.03	90.77	87.84	82	91.67	90.92	73.08	83.63	79.35	72.35
MW-Net2026.03	90.55	88.02	83.45	93.03	91.29	65.42	90.66	82.3	63.83
L2B2026.03	89.15	86.27	80.38	91.37	90.56	89.66	89.67	85.98	74.23
L2B2026.03	88.83	86.27	80.23	91.23	90.51	89.23	89.46	85.79	74.23
FasTEN2026.03	86.19	77.62	70.04	91.63	88.82	85.71	86.82	81.16	61.88
MW-Net2026.03	84.86	72.58	57.04	90.49	87.21	61.65	79.29	66.94	52.41
Cross-Entropy2026.03	82.84	71.3	59.93	88.45	79.92	70.19	78.15	54.54	39.72
EMLC2026.03	81.6	71.24	58.55	88.51	80.95	68.11	80.32	74.09	61.57