# Metaphor Classification on MMSD Metaphor style 2.0 (test)

Source: [Metaphor Classification on MMSD Metaphor style 2.0 (test)](<https://www.sota2.com/research/sota/metaphor-classification-on-mmsd-metaphor-style-2-0-test>)

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

Data updated: 2026-02-26T05:30:58.183Z

Evaluation of Metaphor Classification performance on the test split of the MMSD Metaphor style 2.0 dataset, measured by Accuracy and F1 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 | Metaphor Classification |
| Dataset (including any stated split) | MMSD Metaphor style 2.0 (test) |
| Metric | Accuracy |
| Unit | PERCENT |
| 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

- [Accuracy](<https://www.sota2.com/research/sota/metaphor-classification-on-mmsd-metaphor-style-2-0-test.md?metric=Accuracy>): Higher is better; unit: PERCENT.

- [F1 Score](<https://www.sota2.com/research/sota/metaphor-classification-on-mmsd-metaphor-style-2-0-test.md?metric=F1%20Score>): Higher is better; unit: SCALAR.

## Recorded results

| Method | Accuracy | Result date | Report paper |
| --- | --- | --- | --- |
| Gemini 2.5 Flash | 73.19 | Not recorded | 697846b1a2b28e761d87c5f4 |
| SFT on Combined → GRPO (Combined) | 69.14 | Not recorded | 697846b1a2b28e761d87c5f4 |
| SFT on Combined → GRPO (Style specific) | 68.19 | Not recorded | 697846b1a2b28e761d87c5f4 |
| LLaMA 90B Vision Instruct | 67.56 | Not recorded | 697846b1a2b28e761d87c5f4 |
| Qwen2.5-VL 32B Instruct | 61.31 | Not recorded | 697846b1a2b28e761d87c5f4 |
| Gemma 3 27B | 58.44 | Not recorded | 697846b1a2b28e761d87c5f4 |
| Phi-4 Multimodal | 55.71 | Not recorded | 697846b1a2b28e761d87c5f4 |
| LLaVA 1.5 7B | 53.68 | Not recorded | 697846b1a2b28e761d87c5f4 |

### Gemini 2.5 Flash

Accuracy: 73.19. Result ID: 699670555698665ec4b766f8.

[Reporting paper](<https://www.sota2.com/research/paper/697846b1a2b28e761d87c5f4.md>)

Recorded evaluation configuration: {"Prompting":"zero-shot"}. This record does not assert independent reproduction.

### SFT on Combined → GRPO (Combined)

Accuracy: 69.14. Result ID: 699670555698665ec4b766ff.

[Reporting paper](<https://www.sota2.com/research/paper/697846b1a2b28e761d87c5f4.md>)

[Method origin paper](<https://www.sota2.com/research/paper/697846b1a2b28e761d87c5f4.md>)

Recorded evaluation configuration: {"Prompting":"zero-shot","Setup":"Combined"}. This record does not assert independent reproduction.

### SFT on Combined → GRPO (Style specific)

Accuracy: 68.19. Result ID: 699670555698665ec4b766fe.

[Reporting paper](<https://www.sota2.com/research/paper/697846b1a2b28e761d87c5f4.md>)

[Method origin paper](<https://www.sota2.com/research/paper/697846b1a2b28e761d87c5f4.md>)

Recorded evaluation configuration: {"Prompting":"zero-shot","Setup":"Style specific"}. This record does not assert independent reproduction.

### LLaMA 90B Vision Instruct

Accuracy: 67.56. Result ID: 699670555698665ec4b766fa.

[Reporting paper](<https://www.sota2.com/research/paper/697846b1a2b28e761d87c5f4.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa75d11e209131cea6a028.md>)

[Implementation](<https://github.com/meta-llama/llama-agentic-system>)

Recorded evaluation configuration: {"Prompting":"zero-shot"}. This record does not assert independent reproduction.

### Qwen2.5-VL 32B Instruct

Accuracy: 61.31. Result ID: 699670555698665ec4b766f9.

[Reporting paper](<https://www.sota2.com/research/paper/697846b1a2b28e761d87c5f4.md>)

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

Recorded evaluation configuration: {"Prompting":"zero-shot"}. This record does not assert independent reproduction.

### Gemma 3 27B

Accuracy: 58.44. Result ID: 699670555698665ec4b766fb.

[Reporting paper](<https://www.sota2.com/research/paper/697846b1a2b28e761d87c5f4.md>)

Recorded evaluation configuration: {"Prompting":"zero-shot"}. This record does not assert independent reproduction.

### Phi-4 Multimodal

Accuracy: 55.71. Result ID: 699670555698665ec4b766fc.

[Reporting paper](<https://www.sota2.com/research/paper/697846b1a2b28e761d87c5f4.md>)

Recorded evaluation configuration: {"Prompting":"zero-shot"}. This record does not assert independent reproduction.

### LLaVA 1.5 7B

Accuracy: 53.68. Result ID: 699670555698665ec4b766fd.

[Reporting paper](<https://www.sota2.com/research/paper/697846b1a2b28e761d87c5f4.md>)

Recorded evaluation configuration: {"Prompting":"zero-shot"}. This record does not assert independent reproduction.

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