# Image Similarity Evaluation on 10 diverse text prompts

Source: [Image Similarity Evaluation on 10 diverse text prompts](<https://www.sota2.com/research/sota/image-similarity-evaluation-on-10-diverse-text-prompts>)

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

Data updated: 2026-05-14T08:18:23.142Z

Evaluation of image similarity performance using the '10 diverse text prompts' dataset, specifically measuring baseline performance, attack success rates, and defense efficacy (QRNG Defense).

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 | Image Similarity Evaluation |
| Dataset (including any stated split) | 10 diverse text prompts |
| Metric | Attack Success Rate |
| Unit | PERCENT |
| 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

- [Attack Success Rate](<https://www.sota2.com/research/sota/image-similarity-evaluation-on-10-diverse-text-prompts.md?metric=Attack%20Success%20Rate>): Higher is better; unit: PERCENT.

- [Baseline Score](<https://www.sota2.com/research/sota/image-similarity-evaluation-on-10-diverse-text-prompts.md?metric=Baseline%20Score>): Higher is better; unit: PERCENT.

- [QRNG Defense Score](<https://www.sota2.com/research/sota/image-similarity-evaluation-on-10-diverse-text-prompts.md?metric=QRNG%20Defense%20Score>): Higher is better; unit: PERCENT.

## Recorded results

| Method | Attack Success Rate | Result date | Report paper |
| --- | --- | --- | --- |
| SD 1.4 | 100 | Not recorded | 6a0556f9dbc836b87d828be5 |
| SD 1.5 | 100 | Not recorded | 6a0556f9dbc836b87d828be5 |
| SD 1.5 | 100 | Not recorded | 6a0556f9dbc836b87d828be5 |
| SD 1.5 | 100 | Not recorded | 6a0556f9dbc836b87d828be5 |
| SDXL | 100 | Not recorded | 6a0556f9dbc836b87d828be5 |

### SD 1.4

Attack Success Rate: 100. Result ID: 6a05853d47f34ce924ec6e90.

[Reporting paper](<https://www.sota2.com/research/paper/6a0556f9dbc836b87d828be5.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6c391e209131cea32e4a.md>)

[Implementation](<https://github.com/CompVis/latent-diffusion>)

Recorded evaluation configuration: {"Model":"Stable Diffusion v1.4","η":"0.0","Steps (T)":"50","Guidance Scale (w)":"7.5","Resolution":"512x512"}. This record does not assert independent reproduction.

### SD 1.5

Attack Success Rate: 100. Result ID: 6a05853d47f34ce924ec6e91.

[Reporting paper](<https://www.sota2.com/research/paper/6a0556f9dbc836b87d828be5.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6c391e209131cea32e4a.md>)

[Implementation](<https://github.com/CompVis/latent-diffusion>)

Recorded evaluation configuration: {"Model":"Stable Diffusion v1.5","η":"0.0","Steps (T)":"50","Guidance Scale (w)":"7.5","Resolution":"512x512"}. This record does not assert independent reproduction.

### SD 1.5

Attack Success Rate: 100. Result ID: 6a05853d47f34ce924ec6e92.

[Reporting paper](<https://www.sota2.com/research/paper/6a0556f9dbc836b87d828be5.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6c391e209131cea32e4a.md>)

[Implementation](<https://github.com/CompVis/latent-diffusion>)

Recorded evaluation configuration: {"Model":"Stable Diffusion v1.5","η":"0.2","Steps (T)":"50","Guidance Scale (w)":"7.5","Resolution":"512x512"}. This record does not assert independent reproduction.

### SD 1.5

Attack Success Rate: 100. Result ID: 6a05853d47f34ce924ec6e93.

[Reporting paper](<https://www.sota2.com/research/paper/6a0556f9dbc836b87d828be5.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6c391e209131cea32e4a.md>)

[Implementation](<https://github.com/CompVis/latent-diffusion>)

Recorded evaluation configuration: {"Model":"Stable Diffusion v1.5","η":"0.5","Steps (T)":"50","Guidance Scale (w)":"7.5","Resolution":"512x512"}. This record does not assert independent reproduction.

### SDXL

Attack Success Rate: 100. Result ID: 6a05853d47f34ce924ec6e94.

[Reporting paper](<https://www.sota2.com/research/paper/6a0556f9dbc836b87d828be5.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa70011e209131cea487fb.md>)

Recorded evaluation configuration: {"Model":"Stable Diffusion XL","η":"0.0","Steps (T)":"50","Guidance Scale (w)":"7.5","Resolution":"1024x1024"}. This record does not assert independent reproduction.

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