# Diagram Question Answering on AI2D no mask

Source: [Diagram Question Answering on AI2D no mask](<https://www.sota2.com/research/sota/diagram-question-answering-on-ai2d-no-mask>)

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

Data updated: 2026-02-27T10:00:48.700Z

Evaluation of model performance on the 'no mask' subset of the AI2D dataset for Diagram Question Answering, typically measured by accuracy.

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 | Diagram Question Answering |
| Dataset (including any stated split) | AI2D no mask |
| Metric | Accuracy |
| Unit | PERCENT |
| Direction | Higher is better |
| Results loaded | 7 of 7 |
| 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/diagram-question-answering-on-ai2d-no-mask.md?metric=Accuracy>): Higher is better; unit: PERCENT.

## Recorded results

| Method | Accuracy | Result date | Report paper |
| --- | --- | --- | --- |
| RADIOv2.5-H | 82.5 | Not recorded | 68fa76e01e209131cea6f909 |
| RADIOv2.5-g | 80.8 | Not recorded | 68fa76e01e209131cea6f909 |
| SigLIP-SO400M | 80.7 | Not recorded | 68fa76e01e209131cea6f909 |
| RADIOv2.5-L | 79.2 | Not recorded | 68fa76e01e209131cea6f909 |
| RADIOv2.1-H | 78.9 | Not recorded | 68fa76e01e209131cea6f909 |
| OpenAI-CLIP | 77.6 | Not recorded | 68fa76e01e209131cea6f909 |
| RADIOv2.1-H | 73.4 | Not recorded | 68fa76e01e209131cea6f909 |

### RADIOv2.5-H

Accuracy: 82.5. Result ID: 69a16b3d35dfcabb00e6ec35.

[Reporting paper](<https://www.sota2.com/research/paper/68fa76e01e209131cea6f909.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa76e01e209131cea6f909.md>)

Recorded evaluation configuration: {"Resolution":"768^2","Tokens/im":"196","LLM":"MN-Minitron-8B","Data mixture":"LLaVA1.6"}. This record does not assert independent reproduction.

### RADIOv2.5-g

Accuracy: 80.8. Result ID: 69a16b3d35dfcabb00e6ec36.

[Reporting paper](<https://www.sota2.com/research/paper/68fa76e01e209131cea6f909.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa76e01e209131cea6f909.md>)

Recorded evaluation configuration: {"Resolution":"672^2","Tokens/im":"196","LLM":"MN-Minitron-8B","Data mixture":"LLaVA1.6"}. This record does not assert independent reproduction.

### SigLIP-SO400M

Accuracy: 80.7. Result ID: 69a16b3d35dfcabb00e6ec33.

[Reporting paper](<https://www.sota2.com/research/paper/68fa76e01e209131cea6f909.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa70ae1e209131cea4c7a4.md>)

Recorded evaluation configuration: {"Resolution":"384^2","Tokens/im":"196","LLM":"MN-Minitron-8B","Data mixture":"LLaVA1.6"}. This record does not assert independent reproduction.

### RADIOv2.5-L

Accuracy: 79.2. Result ID: 69a16b3d35dfcabb00e6ec34.

[Reporting paper](<https://www.sota2.com/research/paper/68fa76e01e209131cea6f909.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa76e01e209131cea6f909.md>)

Recorded evaluation configuration: {"Resolution":"768^2","Tokens/im":"196","LLM":"MN-Minitron-8B","Data mixture":"LLaVA1.6"}. This record does not assert independent reproduction.

### RADIOv2.1-H

Accuracy: 78.9. Result ID: 69a16b3d35dfcabb00e6ec31.

[Reporting paper](<https://www.sota2.com/research/paper/68fa76e01e209131cea6f909.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa74541e209131cea611e9.md>)

[Implementation](<https://github.com/NVlabs/RADIO>)

Recorded evaluation configuration: {"Resolution":"512^2","Tokens/im":"196","LLM":"MN-Minitron-8B","Data mixture":"LLaVA1.6"}. This record does not assert independent reproduction.

### OpenAI-CLIP

Accuracy: 77.6. Result ID: 69a16b3d35dfcabb00e6ec30.

[Reporting paper](<https://www.sota2.com/research/paper/68fa76e01e209131cea6f909.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa69581e209131cea21ad0.md>)

[Implementation](<https://github.com/OpenAI/CLIP>)

Recorded evaluation configuration: {"Resolution":"336^2","Tokens/im":"196","LLM":"MN-Minitron-8B","Data mixture":"LLaVA1.6"}. This record does not assert independent reproduction.

### RADIOv2.1-H

Accuracy: 73.4. Result ID: 69a16b3d35dfcabb00e6ec32.

[Reporting paper](<https://www.sota2.com/research/paper/68fa76e01e209131cea6f909.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa74541e209131cea611e9.md>)

[Implementation](<https://github.com/NVlabs/RADIO>)

Recorded evaluation configuration: {"Resolution":"768^2","Tokens/im":"196","LLM":"MN-Minitron-8B","Data mixture":"LLaVA1.6"}. This record does not assert independent reproduction.

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