# Toxicity Detection on DYNAHATE (test)

Source: [Toxicity Detection on DYNAHATE (test)](<https://www.sota2.com/research/sota/toxicity-detection-on-dynahate-test>)

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

Data updated: 2026-02-26T05:24:31.777Z

Evaluation of model performance on the test split of the DYNAHATE dataset for the task of Toxicity Detection, typically reporting AUC.

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 | Toxicity Detection |
| Dataset (including any stated split) | DYNAHATE (test) |
| Metric | AUC |
| Unit | SCALAR |
| Direction | Higher is better |
| Results loaded | 8 of 8 |
| 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

- [AUC](<https://www.sota2.com/research/sota/toxicity-detection-on-dynahate-test.md?metric=AUC>): Higher is better; unit: SCALAR.

## Recorded results

| Method | AUC | Result date | Report paper |
| --- | --- | --- | --- |
| HateBERT | 0.66 | Not recorded | 68fa6d021e209131cea374aa |
| HateBERT | 0.59 | Not recorded | 68fa6d021e209131cea374aa |
| HateBERT | 0.54 | Not recorded | 68fa6d021e209131cea374aa |
| RoBERTa | 0.54 | Not recorded | 68fa6d021e209131cea374aa |
| RoBERTa | 0.51 | Not recorded | 68fa6d021e209131cea374aa |
| RoBERTa | 0.5 | Not recorded | 68fa6d021e209131cea374aa |
| RoBERTa | 0.49 | Not recorded | 68fa6d021e209131cea374aa |
| HateBERT | 0.47 | Not recorded | 68fa6d021e209131cea374aa |

### HateBERT

AUC: 0.66. Result ID: 698f72213a6c314849f8a085.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6d021e209131cea374aa.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6d021e209131cea374aa.md>)

[Implementation](<https://github.com/microsoft/ToxiGen>)

Recorded evaluation configuration: {"Finetune Data":"ALICE + top-k"}. This record does not assert independent reproduction.

### HateBERT

AUC: 0.59. Result ID: 698f72213a6c314849f8a084.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6d021e209131cea374aa.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6d021e209131cea374aa.md>)

[Implementation](<https://github.com/microsoft/ToxiGen>)

Recorded evaluation configuration: {"Finetune Data":"top-k"}. This record does not assert independent reproduction.

### HateBERT

AUC: 0.54. Result ID: 698f72213a6c314849f8a083.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6d021e209131cea374aa.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6d021e209131cea374aa.md>)

[Implementation](<https://github.com/microsoft/ToxiGen>)

Recorded evaluation configuration: {"Finetune Data":"ALICE"}. This record does not assert independent reproduction.

### RoBERTa

AUC: 0.54. Result ID: 698f72213a6c314849f8a089.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6d021e209131cea374aa.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6d021e209131cea374aa.md>)

[Implementation](<https://github.com/microsoft/ToxiGen>)

Recorded evaluation configuration: {"Finetune Data":"ALICE + top-k"}. This record does not assert independent reproduction.

### RoBERTa

AUC: 0.51. Result ID: 698f72213a6c314849f8a087.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6d021e209131cea374aa.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6d021e209131cea374aa.md>)

[Implementation](<https://github.com/microsoft/ToxiGen>)

Recorded evaluation configuration: {"Finetune Data":"ALICE"}. This record does not assert independent reproduction.

### RoBERTa

AUC: 0.5. Result ID: 698f72213a6c314849f8a088.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6d021e209131cea374aa.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6d021e209131cea374aa.md>)

[Implementation](<https://github.com/microsoft/ToxiGen>)

Recorded evaluation configuration: {"Finetune Data":"top-k"}. This record does not assert independent reproduction.

### RoBERTa

AUC: 0.49. Result ID: 698f72213a6c314849f8a086.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6d021e209131cea374aa.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6d021e209131cea374aa.md>)

[Implementation](<https://github.com/microsoft/ToxiGen>)

Recorded evaluation configuration: {"Finetune Data":"None"}. This record does not assert independent reproduction.

### HateBERT

AUC: 0.47. Result ID: 698f72213a6c314849f8a082.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6d021e209131cea374aa.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6d021e209131cea374aa.md>)

[Implementation](<https://github.com/microsoft/ToxiGen>)

Recorded evaluation configuration: {"Finetune Data":"None"}. This record does not assert independent reproduction.

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