# Visual Question Answering on VG VQA Japanese (standard original)

Source: [Visual Question Answering on VG VQA Japanese (standard original)](<https://www.sota2.com/research/sota/visual-question-answering-on-vg-vqa-japanese-standard-original>)

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

Data updated: 2026-02-26T05:27:52.062Z

Evaluation of model performance on the standard original split of the VG VQA Japanese dataset for visual question answering.

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) | VG VQA Japanese (standard original) |
| 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-vg-vqa-japanese-standard-original.md?metric=Accuracy>): Higher is better; unit: PERCENT.

- [BLEU](<https://www.sota2.com/research/sota/visual-question-answering-on-vg-vqa-japanese-standard-original.md?metric=BLEU>): Higher is better; unit: PERCENT.

## Recorded results

| Method | Accuracy | Result date | Report paper |
| --- | --- | --- | --- |
| UC2 | 34.2 | Not recorded | 68fa69b11e209131cea237c4 |
| UNITERcc | 22.7 | Not recorded | 68fa69b11e209131cea237c4 |
| PCATT | 19.2 | Not recorded | 68fa69b11e209131cea237c4 |

### UC2

Accuracy: 34.2. Result ID: 698dd5afffc3901de098f0fc.

[Reporting paper](<https://www.sota2.com/research/paper/68fa69b11e209131cea237c4.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa69b11e209131cea237c4.md>)

Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction.

### UNITERcc

Accuracy: 22.7. Result ID: 698dd5afffc3901de098f0fb.

[Reporting paper](<https://www.sota2.com/research/paper/68fa69b11e209131cea237c4.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa67f91e209131cea1989f.md>)

[Implementation](<https://github.com/ChenRocks/UNITER>)

Recorded evaluation configuration: {"mode":"translate-test baseline"}. This record does not assert independent reproduction.

### PCATT

Accuracy: 19.2. Result ID: 698dd5afffc3901de098f0fa.

[Reporting paper](<https://www.sota2.com/research/paper/68fa69b11e209131cea237c4.md>)

Recorded evaluation configuration: {"split":"different dev/test split"}. This record does not assert independent reproduction.

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