# Content Moderation on OpenAI Out-of-Distribution

Source: [Content Moderation on OpenAI Out-of-Distribution](<https://www.sota2.com/research/sota/content-moderation-on-openai-out-of-distribution>)

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

Data updated: 2026-04-14T09:24:29.253Z

Evaluation of model performance on the OpenAI Out-of-Distribution dataset for Content Moderation, reporting category-specific performance (Pornography, Violence, Bias, Harmlessness) and the average score.

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 | Content Moderation |
| Dataset (including any stated split) | OpenAI Out-of-Distribution |
| Metric | Pornography Score |
| Unit | PERCENT |
| 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

- [Pornography Score](<https://www.sota2.com/research/sota/content-moderation-on-openai-out-of-distribution.md?metric=Pornography%20Score>): Higher is better; unit: PERCENT.

- [Violence Score](<https://www.sota2.com/research/sota/content-moderation-on-openai-out-of-distribution.md?metric=Violence%20Score>): Higher is better; unit: PERCENT.

- [Bias Score](<https://www.sota2.com/research/sota/content-moderation-on-openai-out-of-distribution.md?metric=Bias%20Score>): Higher is better; unit: PERCENT.

- [Harmlessness Score](<https://www.sota2.com/research/sota/content-moderation-on-openai-out-of-distribution.md?metric=Harmlessness%20Score>): Higher is better; unit: PERCENT.

- [Average Score](<https://www.sota2.com/research/sota/content-moderation-on-openai-out-of-distribution.md?metric=Average%20Score>): Higher is better; unit: PERCENT.

## Recorded results

| Method | Pornography Score | Result date | Report paper |
| --- | --- | --- | --- |
| CARO | 82.6 | Not recorded | 69ddca117d0172140ec799b6 |
| Qwen2.5-7B-Instruct | 75.3 | Not recorded | 69ddca117d0172140ec799b6 |

### CARO

Pornography Score: 82.6. Result ID: 69de07c647f34ce924c5be30.

[Reporting paper](<https://www.sota2.com/research/paper/69ddca117d0172140ec799b6.md>)

[Method origin paper](<https://www.sota2.com/research/paper/69ddca117d0172140ec799b6.md>)

Recorded evaluation configuration: {"backbone":"Qwen2.5-7B-Instruct"}. This record does not assert independent reproduction.

### Qwen2.5-7B-Instruct

Pornography Score: 75.3. Result ID: 69de07c647f34ce924c5be2f.

[Reporting paper](<https://www.sota2.com/research/paper/69ddca117d0172140ec799b6.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa767e1e209131cea6d6a7.md>)

Recorded evaluation configuration: {"model\_status":"Base Model"}. This record does not assert independent reproduction.

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