# Science Visual Question Answering on SQAI ScienceQA-Image

Source: [Science Visual Question Answering on SQAI ScienceQA-Image](<https://www.sota2.com/research/sota/science-visual-question-answering-on-sqai-scienceqa-image>)

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

Data updated: 2026-04-13T07:29:35.959Z

Evaluation of model performance on the SQAI ScienceQA-Image subset for Science Visual Question Answering, measured using Accuracy and Relational Accuracy (RelAcc).

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 | Science Visual Question Answering |
| Dataset (including any stated split) | SQAI ScienceQA-Image |
| Metric | Accuracy |
| 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

- [Accuracy](<https://www.sota2.com/research/sota/science-visual-question-answering-on-sqai-scienceqa-image.md?metric=Accuracy>): Higher is better; unit: PERCENT.

- [Relational Accuracy](<https://www.sota2.com/research/sota/science-visual-question-answering-on-sqai-scienceqa-image.md?metric=Relational%20Accuracy>): Higher is better; unit: PERCENT.

## Recorded results

| Method | Accuracy | Result date | Report paper |
| --- | --- | --- | --- |
| LLaVA-1.5-7B | 69.5 | Not recorded | 68fa7a1c1e209131cea81398 |
| TwigVLM++ | 69.5 | Not recorded | 68fa7a1c1e209131cea81398 |
| TwigVLM++ | 69.5 | Not recorded | 68fa7a1c1e209131cea81398 |
| TwigVLM++ | 69.2 | Not recorded | 68fa7a1c1e209131cea81398 |

### LLaVA-1.5-7B

Accuracy: 69.5. Result ID: 69dc9a8e47f34ce924c33965.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7a1c1e209131cea81398.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa732e1e209131cea5af59.md>)

[Implementation](<https://llava-vl.github.io/>)

Recorded evaluation configuration: {"Retained Tokens":"576","Pruning Ratio":"0%","Base Model":"LLaVA-1.5-7B"}. This record does not assert independent reproduction.

### TwigVLM++

Accuracy: 69.5. Result ID: 69dc9a8e47f34ce924c33967.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7a1c1e209131cea81398.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa7a1c1e209131cea81398.md>)

Recorded evaluation configuration: {"Retained Tokens":"128","Pruning Ratio":"77.8%","Base Model":"LLaVA-1.5-7B"}. This record does not assert independent reproduction.

### TwigVLM++

Accuracy: 69.5. Result ID: 69dc9a8e47f34ce924c33968.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7a1c1e209131cea81398.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa7a1c1e209131cea81398.md>)

Recorded evaluation configuration: {"Retained Tokens":"64","Pruning Ratio":"88.9%","Base Model":"LLaVA-1.5-7B"}. This record does not assert independent reproduction.

### TwigVLM++

Accuracy: 69.2. Result ID: 69dc9a8e47f34ce924c33966.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7a1c1e209131cea81398.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa7a1c1e209131cea81398.md>)

Recorded evaluation configuration: {"Retained Tokens":"192","Pruning Ratio":"66.7%","Base Model":"LLaVA-1.5-7B"}. This record does not assert independent reproduction.

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