# Targeted Adversarial Attack on ImageNet (val)

Source: [Targeted Adversarial Attack on ImageNet (val)](<https://www.sota2.com/research/sota/targeted-adversarial-attack-on-imagenet-val>)

Collection index: [llms.txt](<https://www.sota2.com/research/llms.txt>)

Data updated: 2026-02-26T06:08:01.364Z

Evaluation of targeted adversarial attack effectiveness on the ImageNet validation split, reporting performance metrics against a specific suite of 12 target classification models including VGG-16, ResNet-50, ViT, LeViT, and PiT.

A research result applies to its stated dataset, split, metric and evaluation configuration. Compare like-for-like settings, preserve the metric's unit and direction, and distinguish paper-reported results from independent reproduction. The best result recorded here is not a claim of exhaustive, current global coverage.

| Scope | Recorded value |
| --- | --- |
| Task | Targeted Adversarial Attack |
| Dataset (including any stated split) | ImageNet (val) |
| Metric | VGG-16 Performance Score |
| Unit | SCALAR |
| Direction | Higher is better |
| Results loaded | 2 of 2 |
| Coverage | All results currently returned for this benchmark |

Evaluation settings may differ between results. The original papers and recorded configurations are needed to establish comparability. A shared metric name alone does not establish an identical protocol.

## Metrics

- [VGG-16 Performance Score](<https://www.sota2.com/research/sota/targeted-adversarial-attack-on-imagenet-val.md?metric=VGG-16%20Performance%20Score>): Higher is better; unit: SCALAR.

- [ResNet-50 Performance Score](<https://www.sota2.com/research/sota/targeted-adversarial-attack-on-imagenet-val.md?metric=ResNet-50%20Performance%20Score>): Higher is better; unit: SCALAR.

- [Inception-v3 Performance Score](<https://www.sota2.com/research/sota/targeted-adversarial-attack-on-imagenet-val.md?metric=Inception-v3%20Performance%20Score>): Higher is better; unit: SCALAR.

- [DenseNet-121 Performance Score](<https://www.sota2.com/research/sota/targeted-adversarial-attack-on-imagenet-val.md?metric=DenseNet-121%20Performance%20Score>): Higher is better; unit: SCALAR.

- [Inception-ResNet-v2 Performance Score](<https://www.sota2.com/research/sota/targeted-adversarial-attack-on-imagenet-val.md?metric=Inception-ResNet-v2%20Performance%20Score>): Higher is better; unit: SCALAR.

- [Inception-v4 Performance Score](<https://www.sota2.com/research/sota/targeted-adversarial-attack-on-imagenet-val.md?metric=Inception-v4%20Performance%20Score>): Higher is better; unit: SCALAR.

- [Xception Performance Score](<https://www.sota2.com/research/sota/targeted-adversarial-attack-on-imagenet-val.md?metric=Xception%20Performance%20Score>): Higher is better; unit: SCALAR.

- [ViT Performance Score](<https://www.sota2.com/research/sota/targeted-adversarial-attack-on-imagenet-val.md?metric=ViT%20Performance%20Score>): Higher is better; unit: SCALAR.

- [LeViT Performance Score](<https://www.sota2.com/research/sota/targeted-adversarial-attack-on-imagenet-val.md?metric=LeViT%20Performance%20Score>): Higher is better; unit: SCALAR.

- [ConViT Performance Score](<https://www.sota2.com/research/sota/targeted-adversarial-attack-on-imagenet-val.md?metric=ConViT%20Performance%20Score>): Higher is better; unit: SCALAR.

- [Twins Performance Score](<https://www.sota2.com/research/sota/targeted-adversarial-attack-on-imagenet-val.md?metric=Twins%20Performance%20Score>): Higher is better; unit: SCALAR.

- [PiT Performance Score](<https://www.sota2.com/research/sota/targeted-adversarial-attack-on-imagenet-val.md?metric=PiT%20Performance%20Score>): Higher is better; unit: SCALAR.

## Recorded results

| Method | VGG-16 Performance Score | Result date | Report paper |
| --- | --- | --- | --- |
| RDI-FTM-E | 50.1 | Not recorded | 68fa77931e209131cea740d1 |
| DI | 1.6 | Not recorded | 68fa77931e209131cea740d1 |

### RDI-FTM-E

VGG-16 Performance Score: 50.1. Result ID: 699542de3badfaffe62ed699.

[Reporting paper](<https://www.sota2.com/research/paper/68fa77931e209131cea740d1.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa77931e209131cea740d1.md>)

[Implementation](<https://github.com/uhiu/feature-tuning-mixup>)

Recorded evaluation configuration: {"Surrogate Model":"LeViT","Combined with":"MI-TI"}. This record does not assert independent reproduction.

### DI

VGG-16 Performance Score: 1.6. Result ID: 699542de3badfaffe62ed698.

[Reporting paper](<https://www.sota2.com/research/paper/68fa77931e209131cea740d1.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa65dc1e209131cea0ceaa.md>)

Recorded evaluation configuration: {"Surrogate Model":"LeViT","Combined with":"MI-TI"}. This record does not assert independent reproduction.

Page 1 of 1. This is an index page, not the full collection.