# Visual Question Answering on Medical VQA DME dataset (test)

Source: [Visual Question Answering on Medical VQA DME dataset (test)](<https://www.sota2.com/research/sota/visual-question-answering-on-medical-vqa-dme-dataset-test>)

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

Data updated: 2026-02-26T05:54:23.949Z

Evaluation of model performance on the test split of the Medical VQA DME dataset for visual question answering capabilities.

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 | Visual Question Answering |
| Dataset (including any stated split) | Medical VQA DME dataset (test) |
| Metric | Overall Accuracy |
| Unit | SCALAR |
| 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

- [Overall Accuracy](<https://www.sota2.com/research/sota/visual-question-answering-on-medical-vqa-dme-dataset-test.md?metric=Overall%20Accuracy>): Higher is better; unit: SCALAR.

- [Grade Accuracy](<https://www.sota2.com/research/sota/visual-question-answering-on-medical-vqa-dme-dataset-test.md?metric=Grade%20Accuracy>): Higher is better; unit: SCALAR.

- [Whole Accuracy](<https://www.sota2.com/research/sota/visual-question-answering-on-medical-vqa-dme-dataset-test.md?metric=Whole%20Accuracy>): Higher is better; unit: SCALAR.

- [Macula Accuracy](<https://www.sota2.com/research/sota/visual-question-answering-on-medical-vqa-dme-dataset-test.md?metric=Macula%20Accuracy>): Higher is better; unit: SCALAR.

- [Region Accuracy](<https://www.sota2.com/research/sota/visual-question-answering-on-medical-vqa-dme-dataset-test.md?metric=Region%20Accuracy>): Higher is better; unit: SCALAR.

- [C1 Consistency](<https://www.sota2.com/research/sota/visual-question-answering-on-medical-vqa-dme-dataset-test.md?metric=C1%20Consistency>): Higher is better; unit: SCALAR.

- [C2 Consistency](<https://www.sota2.com/research/sota/visual-question-answering-on-medical-vqa-dme-dataset-test.md?metric=C2%20Consistency>): Higher is better; unit: SCALAR.

## Recorded results

| Method | Overall Accuracy | Result date | Report paper |
| --- | --- | --- | --- |
| Consistency-preserving VQA | 0.8349 | Not recorded | 68fa6ce21e209131cea36844 |
| Baseline | 0.8146 | Not recorded | 68fa6ce21e209131cea36844 |
| SQUINT | 0.8058 | Not recorded | 68fa6ce21e209131cea36844 |
| Baseline | 0.7754 | Not recorded | 68fa6ce21e209131cea36844 |

### Consistency-preserving VQA

Overall Accuracy: 0.8349. Result ID: 698f49223a6c314849f7ade6.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6ce21e209131cea36844.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6ce21e209131cea36844.md>)

[Implementation](<https://github.com/sergiotasconmorales/consistency_vqa>)

Recorded evaluation configuration: {"baseline":"Baseline (att.)","lambda":0.5,"gamma":1}. This record does not assert independent reproduction.

### Baseline

Overall Accuracy: 0.8146. Result ID: 698f49223a6c314849f7ade4.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6ce21e209131cea36844.md>)

Recorded evaluation configuration: {"attention":"yes"}. This record does not assert independent reproduction.

### SQUINT

Overall Accuracy: 0.8058. Result ID: 698f49223a6c314849f7ade5.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6ce21e209131cea36844.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa67b41e209131cea17f10.md>)

Recorded evaluation configuration: {"baseline":"Baseline (att.)"}. This record does not assert independent reproduction.

### Baseline

Overall Accuracy: 0.7754. Result ID: 698f49223a6c314849f7ade3.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6ce21e209131cea36844.md>)

Recorded evaluation configuration: {"attention":"none"}. This record does not assert independent reproduction.

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