Adversarial Attack Perturbation Analysis on ImageNet benchmark leaderboard · SOTA2 Research

Adversarial Attack Perturbation Analysis on ImageNet

72.04ParsimonPyTorch ResNet-50

Evaluation Results

Method	Links
PyTorch ResNet-502025.11		72.04	84.64	80.8	55.8	87.28	2.64
Salman ResNet-182025.11		58.6	58.72	54.92	54.26	61.98	3.26
Wong ResNet-502025.11		56.62	56.62	52.46	50.34	59.96	3.34
Engstrom ResNet-502025.11		48.92	48.16	45.1	44.12	51.84	2.92
Salman ResNet-502025.11		46.96	46.7	44.06	43.08	50.16	3.2
Salman WideResNet-50-22025.11		42.94	42.22	39.66	38.32	44.86	1.92
Debenedetti XCiT-S122025.11		37.44	36.48	33.74	32.96	39.66	2.22
Debenedetti XCiT-M122025.11		36.04	35.1	32.7	31.86	37.64	1.6
Singh ViT-S+ConvStem2025.11		35.5	35.3	32.68	32.44	37.58	2.08
Debenedetti XCiT-L122025.11		35.32	34.64	32.38	31.52	37.02	1.7
Singh ConvNeXt-T+ConvStem2025.11		34.88	34.5	32.24	31.58	37.12	2.24
Singh ConvNeXt-S+ConvStem2025.11		33.36	32.8	30.94	30.62	35.28	1.92
Singh ViT-B+ConvStem2025.11		31.02	30.38	28.76	28.2	32.56	1.54
Liu Swin-B2025.11		30.88	30.44	28.64	28.08	32.6	1.72
Singh ConvNeXt-B+ConvStem2025.11		30.78	30.14	28.62	28.24	32.44	1.66
Liu ConvNeXt-B2025.11		30.32	29.76	28.22	27.62	31.94	1.62
Singh ConvNeXt-L+ConvStem2025.11		29.24	28.8	27.3	26.42	30.64	1.4
Liu ConvNeXt-L2025.11		28.92	28.34	26.86	26.4	30.2	1.28
Liu Swin-L2025.11		28.18	27.44	25.86	25.32	29.9	1.72