Adversarial Attack on ImageNet (val) (Specific Model Set)

0.999ViT-B ScoreGNS

Evaluation results

Method
GNS2025.03	0.999	0.895	0.681	0.74	0.824	0.936	0.722	0.778	0.767	0.76	0.791	0.76	0.62	0.773
FPR2025.03	0.992	0.925	0.73	0.783	0.885	0.982	0.777	0.873	0.876	0.82	0.903	0.874	0.683	0.843
PNAPO2025.03	0.991	0.832	0.621	0.658	0.747	0.83	0.64	0.674	0.7	0.676	0.685	0.63	0.563	0.688
TGR2025.03	0.991	0.862	0.638	0.699	0.811	0.944	0.704	0.791	0.798	0.753	0.831	0.799	0.6	0.769
MIM2025.03	0.972	0.618	0.404	0.423	0.549	0.658	0.444	0.517	0.572	0.513	0.508	0.517	0.382	0.509
FPR2025.03	0.946	0.984	0.838	0.867	0.955	0.995	0.87	0.898	0.892	0.881	0.939	0.868	0.798	0.902
FPR2025.03	0.938	1	0.816	0.856	0.95	0.988	0.868	0.874	0.887	0.886	0.906	0.829	0.768	0.881
TGR2025.03	0.924	0.971	0.812	0.843	0.926	0.979	0.835	0.86	0.852	0.85	0.899	0.845	0.746	0.872
GNS2025.03	0.921	0.98	0.805	0.842	0.91	0.98	0.798	0.833	0.807	0.835	0.867	0.813	0.708	0.854
GNS2025.03	0.912	1	0.786	0.822	0.916	0.976	0.811	0.866	0.871	0.849	0.896	0.819	0.708	0.853
TGR2025.03	0.901	1	0.76	0.813	0.912	0.978	0.806	0.869	0.866	0.843	0.877	0.821	0.691	0.845
MIM2025.03	0.86	0.824	0.67	0.695	0.797	0.877	0.653	0.69	0.705	0.701	0.714	0.614	0.582	0.722
PNAPO2025.03	0.834	0.877	0.724	0.741	0.822	0.86	0.719	0.749	0.747	0.749	0.761	0.68	0.663	0.764
PNAPO2025.03	0.698	0.882	0.581	0.61	0.703	0.738	0.601	0.648	0.697	0.643	0.655	0.617	0.532	0.644
MIM2025.03	0.685	0.988	0.5	0.549	0.682	0.753	0.548	0.61	0.678	0.598	0.61	0.557	0.471	0.603
FPR2025.03	0.443	0.561	0.571	0.624	0.695	0.913	0.722	0.887	0.864	0.804	0.876	0.911	0.603	0.729
PNAPO2025.03	0.395	0.509	0.496	0.549	0.616	0.854	0.65	0.787	0.767	0.722	0.779	0.81	0.508	0.649
TGR2025.03	0.379	0.477	0.469	0.541	0.625	0.874	0.651	0.828	0.82	0.755	0.818	0.888	0.537	0.666
GNS2025.03	0.371	0.441	0.464	0.519	0.594	0.844	0.633	0.799	0.78	0.715	0.763	0.843	0.486	0.635
MIM2025.03	0.3	0.361	0.385	0.433	0.51	0.75	0.555	0.695	0.687	0.608	0.645	0.716	0.404	0.542