# Video Classification on HateMM English (test)

Source: [Video Classification on HateMM English (test)](<https://www.sota2.com/research/sota/video-classification-on-hatemm-english-test>)

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

Data updated: 2026-02-26T05:31:32.549Z

Evaluation of Video Classification performance on the test split of the HateMM English dataset, reporting Accuracy, Macro-F1, and class-specific metrics for the hate class.

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 | Video Classification |
| Dataset (including any stated split) | HateMM English (test) |
| Metric | Accuracy |
| Unit | PERCENT |
| Direction | Higher is better |
| Results loaded | 15 of 15 |
| 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

- [Accuracy](<https://www.sota2.com/research/sota/video-classification-on-hatemm-english-test.md?metric=Accuracy>): Higher is better; unit: PERCENT.

- [Macro F1](<https://www.sota2.com/research/sota/video-classification-on-hatemm-english-test.md?metric=Macro%20F1>): Higher is better; unit: PERCENT.

- [F1 (hate class)](<https://www.sota2.com/research/sota/video-classification-on-hatemm-english-test.md?metric=F1%20(hate%20class)>): Higher is better; unit: PERCENT.

- [Precision (hate class)](<https://www.sota2.com/research/sota/video-classification-on-hatemm-english-test.md?metric=Precision%20(hate%20class)>): Higher is better; unit: PERCENT.

- [Recall (hate class)](<https://www.sota2.com/research/sota/video-classification-on-hatemm-english-test.md?metric=Recall%20(hate%20class)>): Higher is better; unit: PERCENT.

## Recorded results

| Method | Accuracy | Result date | Report paper |
| --- | --- | --- | --- |
| MoRE | 82.1 | Not recorded | 6971af35e25ad2fa9d1edb9b |
| CMFusion | 80.2 | Not recorded | 6971af35e25ad2fa9d1edb9b |
| Baseline | 79.8 | Not recorded | 6971af35e25ad2fa9d1edb9b |
| MARS | 78.4 | Not recorded | 6971af35e25ad2fa9d1edb9b |
| MARS | 75.8 | Not recorded | 6971af35e25ad2fa9d1edb9b |
| MARS | 75.1 | Not recorded | 6971af35e25ad2fa9d1edb9b |
| MARS | 72.3 | Not recorded | 6971af35e25ad2fa9d1edb9b |
| Qwen2.5 | 72.1 | Not recorded | 6971af35e25ad2fa9d1edb9b |
| Qwen2.5 | 70.3 | Not recorded | 6971af35e25ad2fa9d1edb9b |
| LLaMA4 | 69.8 | Not recorded | 6971af35e25ad2fa9d1edb9b |
| GPT5 | 68.4 | Not recorded | 6971af35e25ad2fa9d1edb9b |
| GPT5 | 68.4 | Not recorded | 6971af35e25ad2fa9d1edb9b |
| LLaMA4 | 66.9 | Not recorded | 6971af35e25ad2fa9d1edb9b |
| Gemini2.5 | 62.7 | Not recorded | 6971af35e25ad2fa9d1edb9b |
| Gemini2.5 | 60.1 | Not recorded | 6971af35e25ad2fa9d1edb9b |

### MoRE

Accuracy: 82.1. Result ID: 69979e3f5698665ec4f436db.

[Reporting paper](<https://www.sota2.com/research/paper/6971af35e25ad2fa9d1edb9b.md>)

Recorded evaluation configuration: {"Training Paradigm":"Training Based"}. This record does not assert independent reproduction.

### CMFusion

Accuracy: 80.2. Result ID: 69979e3f5698665ec4f436da.

[Reporting paper](<https://www.sota2.com/research/paper/6971af35e25ad2fa9d1edb9b.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa78991e209131cea79366.md>)

Recorded evaluation configuration: {"Training Paradigm":"Training Based"}. This record does not assert independent reproduction.

### Baseline

Accuracy: 79.8. Result ID: 69979e3f5698665ec4f436d9.

[Reporting paper](<https://www.sota2.com/research/paper/6971af35e25ad2fa9d1edb9b.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6f4b1e209131cea44db5.md>)

[Implementation](<https://github.com/hate-alert/HateMM>)

Recorded evaluation configuration: {"Training Paradigm":"Training Based"}. This record does not assert independent reproduction.

### MARS

Accuracy: 78.4. Result ID: 69979e3f5698665ec4f436e5.

[Reporting paper](<https://www.sota2.com/research/paper/6971af35e25ad2fa9d1edb9b.md>)

[Method origin paper](<https://www.sota2.com/research/paper/6971af35e25ad2fa9d1edb9b.md>)

Recorded evaluation configuration: {"Training Paradigm":"Training Free","Prompting Strategy":"MARS","Backbone":"LLaMA4"}. This record does not assert independent reproduction.

### MARS

Accuracy: 75.8. Result ID: 69979e3f5698665ec4f436e4.

[Reporting paper](<https://www.sota2.com/research/paper/6971af35e25ad2fa9d1edb9b.md>)

[Method origin paper](<https://www.sota2.com/research/paper/6971af35e25ad2fa9d1edb9b.md>)

Recorded evaluation configuration: {"Training Paradigm":"Training Free","Prompting Strategy":"MARS","Backbone":"Qwen2.5"}. This record does not assert independent reproduction.

### MARS

Accuracy: 75.1. Result ID: 69979e3f5698665ec4f436e6.

[Reporting paper](<https://www.sota2.com/research/paper/6971af35e25ad2fa9d1edb9b.md>)

[Method origin paper](<https://www.sota2.com/research/paper/6971af35e25ad2fa9d1edb9b.md>)

Recorded evaluation configuration: {"Training Paradigm":"Training Free","Prompting Strategy":"MARS","Backbone":"GPT5"}. This record does not assert independent reproduction.

### MARS

Accuracy: 72.3. Result ID: 69979e3f5698665ec4f436e7.

[Reporting paper](<https://www.sota2.com/research/paper/6971af35e25ad2fa9d1edb9b.md>)

[Method origin paper](<https://www.sota2.com/research/paper/6971af35e25ad2fa9d1edb9b.md>)

Recorded evaluation configuration: {"Training Paradigm":"Training Free","Prompting Strategy":"MARS","Backbone":"Gemini2.5"}. This record does not assert independent reproduction.

### Qwen2.5

Accuracy: 72.1. Result ID: 69979e3f5698665ec4f436e0.

[Reporting paper](<https://www.sota2.com/research/paper/6971af35e25ad2fa9d1edb9b.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa773a1e209131cea71d3c.md>)

Recorded evaluation configuration: {"Training Paradigm":"Training Free","Prompting Strategy":"Chain-of-Thought Prompt"}. This record does not assert independent reproduction.

### Qwen2.5

Accuracy: 70.3. Result ID: 69979e3f5698665ec4f436dc.

[Reporting paper](<https://www.sota2.com/research/paper/6971af35e25ad2fa9d1edb9b.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa773a1e209131cea71d3c.md>)

Recorded evaluation configuration: {"Training Paradigm":"Training Free","Prompting Strategy":"Simple Prompt"}. This record does not assert independent reproduction.

### LLaMA4

Accuracy: 69.8. Result ID: 69979e3f5698665ec4f436dd.

[Reporting paper](<https://www.sota2.com/research/paper/6971af35e25ad2fa9d1edb9b.md>)

Recorded evaluation configuration: {"Training Paradigm":"Training Free","Prompting Strategy":"Simple Prompt"}. This record does not assert independent reproduction.

### GPT5

Accuracy: 68.4. Result ID: 69979e3f5698665ec4f436de.

[Reporting paper](<https://www.sota2.com/research/paper/6971af35e25ad2fa9d1edb9b.md>)

Recorded evaluation configuration: {"Training Paradigm":"Training Free","Prompting Strategy":"Simple Prompt"}. This record does not assert independent reproduction.

### GPT5

Accuracy: 68.4. Result ID: 69979e3f5698665ec4f436e2.

[Reporting paper](<https://www.sota2.com/research/paper/6971af35e25ad2fa9d1edb9b.md>)

Recorded evaluation configuration: {"Training Paradigm":"Training Free","Prompting Strategy":"Chain-of-Thought Prompt"}. This record does not assert independent reproduction.

### LLaMA4

Accuracy: 66.9. Result ID: 69979e3f5698665ec4f436e1.

[Reporting paper](<https://www.sota2.com/research/paper/6971af35e25ad2fa9d1edb9b.md>)

Recorded evaluation configuration: {"Training Paradigm":"Training Free","Prompting Strategy":"Chain-of-Thought Prompt"}. This record does not assert independent reproduction.

### Gemini2.5

Accuracy: 62.7. Result ID: 69979e3f5698665ec4f436df.

[Reporting paper](<https://www.sota2.com/research/paper/6971af35e25ad2fa9d1edb9b.md>)

Recorded evaluation configuration: {"Training Paradigm":"Training Free","Prompting Strategy":"Simple Prompt"}. This record does not assert independent reproduction.

### Gemini2.5

Accuracy: 60.1. Result ID: 69979e3f5698665ec4f436e3.

[Reporting paper](<https://www.sota2.com/research/paper/6971af35e25ad2fa9d1edb9b.md>)

Recorded evaluation configuration: {"Training Paradigm":"Training Free","Prompting Strategy":"Chain-of-Thought Prompt"}. This record does not assert independent reproduction.

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