# Hate speech score reconstruction on MHS small models

Source: [Hate speech score reconstruction on MHS small models](<https://www.sota2.com/research/sota/hate-speech-score-reconstruction-on-mhs-small-models>)

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

Data updated: 2026-05-27T09:52:18.344Z

Evaluation of model performance on the MHS small models dataset for the task of Hate Speech Score Reconstruction, measured using R².

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 | Hate speech score reconstruction |
| Dataset (including any stated split) | MHS small models |
| Metric | R² |
| 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

- [R²](<https://www.sota2.com/research/sota/hate-speech-score-reconstruction-on-mhs-small-models.md?metric=R%C2%B2>): Higher is better; unit: PERCENT.

## Recorded results

| Method | R² | Result date | Report paper |
| --- | --- | --- | --- |
| Vanilla | 62.55 | Not recorded | 6a167acadbd7150506f801e1 |
| Persona | 61.03 | Not recorded | 6a167acadbd7150506f801e1 |
| Vanilla | 60.53 | Not recorded | 6a167acadbd7150506f801e1 |
| Persona | 57.92 | Not recorded | 6a167acadbd7150506f801e1 |

### Vanilla

R²: 62.55. Result ID: 6a16bebffbc373592c105d30.

[Reporting paper](<https://www.sota2.com/research/paper/6a167acadbd7150506f801e1.md>)

[Method origin paper](<https://www.sota2.com/research/paper/6a167acadbd7150506f801e1.md>)

Recorded evaluation configuration: {"Backbone":"Llama-8B"}. This record does not assert independent reproduction.

### Persona

R²: 61.03. Result ID: 6a16bebffbc373592c105d31.

[Reporting paper](<https://www.sota2.com/research/paper/6a167acadbd7150506f801e1.md>)

[Method origin paper](<https://www.sota2.com/research/paper/6a167acadbd7150506f801e1.md>)

Recorded evaluation configuration: {"Backbone":"Llama-8B"}. This record does not assert independent reproduction.

### Vanilla

R²: 60.53. Result ID: 6a16bebffbc373592c105d32.

[Reporting paper](<https://www.sota2.com/research/paper/6a167acadbd7150506f801e1.md>)

[Method origin paper](<https://www.sota2.com/research/paper/6a167acadbd7150506f801e1.md>)

Recorded evaluation configuration: {"Backbone":"Qwen-7B"}. This record does not assert independent reproduction.

### Persona

R²: 57.92. Result ID: 6a16bebffbc373592c105d33.

[Reporting paper](<https://www.sota2.com/research/paper/6a167acadbd7150506f801e1.md>)

[Method origin paper](<https://www.sota2.com/research/paper/6a167acadbd7150506f801e1.md>)

Recorded evaluation configuration: {"Backbone":"Qwen-7B"}. This record does not assert independent reproduction.

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