# Explanation Generation on Hateful Memes (test)

Source: [Explanation Generation on Hateful Memes (test)](<https://www.sota2.com/research/sota/explanation-generation-on-hateful-memes-test>)

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

Data updated: 2026-06-16T08:09:33.550Z

Evaluation of model performance on generating textual explanations for memes using the test split of the Hateful Memes dataset.

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 | Explanation Generation |
| Dataset (including any stated split) | Hateful Memes (test) |
| Metric | BS |
| Unit | PERCENT |
| Direction | Higher is better |
| Results loaded | 3 of 3 |
| 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

- [BS](<https://www.sota2.com/research/sota/explanation-generation-on-hateful-memes-test.md?metric=BS>): Higher is better; unit: PERCENT.

- [MET](<https://www.sota2.com/research/sota/explanation-generation-on-hateful-memes-test.md?metric=MET>): Higher is better; unit: PERCENT.

## Recorded results

| Method | BS | Result date | Report paper |
| --- | --- | --- | --- |
| \[43\] | 78 | Not recorded | 6a30d8c1d33dfb23e783cdde |
| Supervised GRPO | 78 | Not recorded | 6a30d8c1d33dfb23e783cdde |
| Self-supervised GRPO | 77 | Not recorded | 6a30d8c1d33dfb23e783cdde |

### \[43\]

BS: 78. Result ID: 6a3103dffbc373592c2d1aaa.

[Reporting paper](<https://www.sota2.com/research/paper/6a30d8c1d33dfb23e783cdde.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa7ba11e209131cea89724.md>)

Recorded evaluation configuration: {"Category":"Prior work"}. This record does not assert independent reproduction.

### Supervised GRPO

BS: 78. Result ID: 6a3103dffbc373592c2d1aab.

[Reporting paper](<https://www.sota2.com/research/paper/6a30d8c1d33dfb23e783cdde.md>)

[Method origin paper](<https://www.sota2.com/research/paper/6a30d8c1d33dfb23e783cdde.md>)

Recorded evaluation configuration: {"Category":"Proposed","Training":"Supervised"}. This record does not assert independent reproduction.

### Self-supervised GRPO

BS: 77. Result ID: 6a3103dffbc373592c2d1aac.

[Reporting paper](<https://www.sota2.com/research/paper/6a30d8c1d33dfb23e783cdde.md>)

[Method origin paper](<https://www.sota2.com/research/paper/6a30d8c1d33dfb23e783cdde.md>)

Recorded evaluation configuration: {"Category":"Proposed","Training":"Self-supervised"}. This record does not assert independent reproduction.

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