Image Classification on ImageNet (Accuracy, ∆CA, ASR)
13.8Delta CAPRISM
Evaluation results
| Method | Links | |||
|---|---|---|---|---|
| PRISMIR=1, Attack Type=BPP, Backbone=ResNet-502026.01 | 13.8 | — | 2.5 | |
| PRISMIR=1, Attack Type=Blend, Backbone=ResNet-502026.01 | 9.9 | — | 0 | |
| PRISMIR=1, Attack Type=CTRL, Backbone=ResNet-502026.01 | 5.4 | — | 0 | |
| RN50 BaselineIR=1, Backbone=ResNet-502026.01 | — | 49.3 | — | |
| VLM BaselineIR=1, Backbone=VLM2026.01 | — | 59.6 | — |