# Targeted Adversarial Attack on ImageNet-compatible 100 images

Source: [Targeted Adversarial Attack on ImageNet-compatible 100 images](<https://www.sota2.com/research/sota/targeted-adversarial-attack-on-imagenet-compatible-100-images>)

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

Data updated: 2026-02-26T06:04:54.368Z

Evaluation of targeted adversarial attack effectiveness using a specific 100-image subset of the ImageNet-Compatible dataset, reporting outcome counts such as success, failure, refusal, and uncertainty.

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-compatible 100 images |
| Metric | Success Count |
| Unit | SCALAR |
| Direction | Higher is better |
| Results loaded | 5 of 5 |
| 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

- [Success Count](<https://www.sota2.com/research/sota/targeted-adversarial-attack-on-imagenet-compatible-100-images.md?metric=Success%20Count>): Higher is better; unit: SCALAR.

- [Failure Count](<https://www.sota2.com/research/sota/targeted-adversarial-attack-on-imagenet-compatible-100-images.md?metric=Failure%20Count>): Lower is better; unit: SCALAR.

- [Refusal Count](<https://www.sota2.com/research/sota/targeted-adversarial-attack-on-imagenet-compatible-100-images.md?metric=Refusal%20Count>): Lower is better; unit: SCALAR.

- [Uncertainty Count](<https://www.sota2.com/research/sota/targeted-adversarial-attack-on-imagenet-compatible-100-images.md?metric=Uncertainty%20Count>): Lower is better; unit: SCALAR.

## Recorded results

| Method | Success Count | Result date | Report paper |
| --- | --- | --- | --- |
| RDI-FTM-E | 47 | Not recorded | 68fa77931e209131cea740d1 |
| RDI-FTM-E | 45 | Not recorded | 68fa77931e209131cea740d1 |
| RDI-FTM-E | 43 | Not recorded | 68fa77931e209131cea740d1 |
| RDI-FTM-E | 40.5 | Not recorded | 68fa77931e209131cea740d1 |
| RDI-FTM-E | 27 | Not recorded | 68fa77931e209131cea740d1 |

### RDI-FTM-E

Success Count: 47. Result ID: 699542de3badfaffe62ed680.

[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: {"Target Model":"Qwen2-VL","Surrogate Model":"ViT","Prompt":"Is this image a photo of {target label}? Yes or No?"}. This record does not assert independent reproduction.

### RDI-FTM-E

Success Count: 45. Result ID: 699542de3badfaffe62ed682.

[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: {"Target Model":"Claude-3.5","Surrogate Model":"ViT","Prompt":"Is this image a photo of {target label}? Yes or No?"}. This record does not assert independent reproduction.

### RDI-FTM-E

Success Count: 43. Result ID: 699542de3badfaffe62ed681.

[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: {"Target Model":"Llama-3.2","Surrogate Model":"ViT","Prompt":"Is this image a photo of {target label}? Yes or No?"}. This record does not assert independent reproduction.

### RDI-FTM-E

Success Count: 40.5. Result ID: 699542de3badfaffe62ed684.

[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: {"Target Model":"Average","Surrogate Model":"ViT","Prompt":"Is this image a photo of {target label}? Yes or No?"}. This record does not assert independent reproduction.

### RDI-FTM-E

Success Count: 27. Result ID: 699542de3badfaffe62ed683.

[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: {"Target Model":"GPT-4o","Surrogate Model":"ViT","Prompt":"Is this image a photo of {target label}? Yes or No?"}. This record does not assert independent reproduction.

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