# Unsafe Prompt Detection on XSTest

Source: [Unsafe Prompt Detection on XSTest](<https://www.sota2.com/research/sota/unsafe-prompt-detection-on-xstest>)

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

Data updated: 2026-02-26T05:25:39.575Z

Evaluation of model performance on detecting unsafe prompts using the XSTest dataset, reporting Area Under the Precision-Recall Curve (AUPRC).

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 | Unsafe prompt detection |
| Dataset (including any stated split) | XSTest |
| Metric | AUPRC |
| Unit | PERCENT |
| Direction | Higher is better |
| Results loaded | 4 of 4 |
| 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

- [AUPRC](<https://www.sota2.com/research/sota/unsafe-prompt-detection-on-xstest.md?metric=AUPRC>): Higher is better; unit: PERCENT.

## Recorded results

| Method | AUPRC | Result date | Report paper |
| --- | --- | --- | --- |
| GradSafe-Zero | 93.6 | Not recorded | 68fa736c1e209131cea5c676 |
| Llama Guard | 88.9 | Not recorded | 68fa736c1e209131cea5c676 |
| OpenAI Moderation API | 77.9 | Not recorded | 68fa736c1e209131cea5c676 |
| Perspective API | 71.3 | Not recorded | 68fa736c1e209131cea5c676 |

### GradSafe-Zero

AUPRC: 93.6. Result ID: 6992ea2a3a6c31484912561b.

[Reporting paper](<https://www.sota2.com/research/paper/68fa736c1e209131cea5c676.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa736c1e209131cea5c676.md>)

[Implementation](<https://github.com/xyq7/GradSafe>)

Recorded evaluation configuration: {"Backbone":"Llama-2-7b-chat-hf"}. This record does not assert independent reproduction.

### Llama Guard

AUPRC: 88.9. Result ID: 6992ea2a3a6c31484912561a.

[Reporting paper](<https://www.sota2.com/research/paper/68fa736c1e209131cea5c676.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa717b1e209131cea5186d.md>)

Recorded evaluation configuration: {"Backbone":"Llama-2 7b"}. This record does not assert independent reproduction.

### OpenAI Moderation API

AUPRC: 77.9. Result ID: 6992ea2a3a6c314849125618.

[Reporting paper](<https://www.sota2.com/research/paper/68fa736c1e209131cea5c676.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6e8b1e209131cea40729.md>)

Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction.

### Perspective API

AUPRC: 71.3. Result ID: 6992ea2a3a6c314849125619.

[Reporting paper](<https://www.sota2.com/research/paper/68fa736c1e209131cea5c676.md>)

Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction.

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