# Visual Question Answering on CoGenT (TestB)

Source: [Visual Question Answering on CoGenT (TestB)](<https://www.sota2.com/research/sota/visual-question-answering-on-cogent-testb>)

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

Data updated: 2026-02-26T05:28:00.328Z

Evaluation of model performance on the TestB split of the CoGenT dataset for Visual 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 | Visual Question Answering |
| Dataset (including any stated split) | CoGenT (TestB) |
| Metric | Accuracy |
| Unit | PERCENT |
| Direction | Higher is better |
| Results loaded | 3 of 3 |
| 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/visual-question-answering-on-cogent-testb.md?metric=Accuracy>): Higher is better; unit: PERCENT.

## Recorded results

| Method | Accuracy | Result date | Report paper |
| --- | --- | --- | --- |
| FiLM | 78.8 | Not recorded | 68fa6af51e209131cea2b5a9 |
| MDETR | 76.7 | Not recorded | 68fa6af51e209131cea2b5a9 |
| NS-VQA | 63.9 | Not recorded | 68fa6af51e209131cea2b5a9 |

### FiLM

Accuracy: 78.8. Result ID: 698e474effc3901de09c41e1.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6af51e209131cea2b5a9.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa63ee1e209131cea0190e.md>)

[Implementation](<https://github.com/ethanjperez/film>)

Recorded evaluation configuration: {"training condition":"A","fine-tuned on condition B":false}. This record does not assert independent reproduction.

### MDETR

Accuracy: 76.7. Result ID: 698e474effc3901de09c41e3.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6af51e209131cea2b5a9.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6af51e209131cea2b5a9.md>)

[Implementation](<https://github.com/ashkamath/mdetr>)

Recorded evaluation configuration: {"training condition":"A","fine-tuned on condition B":false}. This record does not assert independent reproduction.

### NS-VQA

Accuracy: 63.9. Result ID: 698e474effc3901de09c41e2.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6af51e209131cea2b5a9.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa654e1e209131cea097b7.md>)

[Implementation](<https://github.com/kexinyi/ns-vqa>)

Recorded evaluation configuration: {"training condition":"A","fine-tuned on condition B":false,"external program annotations":true}. This record does not assert independent reproduction.

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