# Explanatory Visual Question Answering on GQA-REX

Source: [Explanatory Visual Question Answering on GQA-REX](<https://www.sota2.com/research/sota/explanatory-visual-question-answering-on-gqa-rex>)

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

Data updated: 2026-05-06T08:35:34.189Z

Evaluation of model performance on the GQA-REX dataset for Explanatory Visual Question Answering, reporting component scores for Consistency (Con.), Visual grounding (Vis.), Textual grounding (Tex.), and overall Average (Ave.).

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 | Explanatory Visual Question Answering |
| Dataset (including any stated split) | GQA-REX |
| Metric | Consistency Score |
| 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

- [Consistency Score](<https://www.sota2.com/research/sota/explanatory-visual-question-answering-on-gqa-rex.md?metric=Consistency%20Score>): Higher is better; unit: PERCENT.

- [Visual Grounding Score](<https://www.sota2.com/research/sota/explanatory-visual-question-answering-on-gqa-rex.md?metric=Visual%20Grounding%20Score>): Higher is better; unit: SCALAR.

- [Textual Grounding Score](<https://www.sota2.com/research/sota/explanatory-visual-question-answering-on-gqa-rex.md?metric=Textual%20Grounding%20Score>): Higher is better; unit: SCALAR.

- [Average Score](<https://www.sota2.com/research/sota/explanatory-visual-question-answering-on-gqa-rex.md?metric=Average%20Score>): Higher is better; unit: SCALAR.

## Recorded results

| Method | Consistency Score | Result date | Report paper |
| --- | --- | --- | --- |
| CgRAG | 97 | Not recorded | 69facb17c0f56ed84308a3f6 |
| VCIN | 93.4 | Not recorded | 69facb17c0f56ed84308a3f6 |
| REX | 84.8 | Not recorded | 69facb17c0f56ed84308a3f6 |

### CgRAG

Consistency Score: 97. Result ID: 69fafcf247f34ce924dded96.

[Reporting paper](<https://www.sota2.com/research/paper/69facb17c0f56ed84308a3f6.md>)

[Method origin paper](<https://www.sota2.com/research/paper/69facb17c0f56ed84308a3f6.md>)

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

### VCIN

Consistency Score: 93.4. Result ID: 69fafcf247f34ce924dded95.

[Reporting paper](<https://www.sota2.com/research/paper/69facb17c0f56ed84308a3f6.md>)

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

### REX

Consistency Score: 84.8. Result ID: 69fafcf247f34ce924dded94.

[Reporting paper](<https://www.sota2.com/research/paper/69facb17c0f56ed84308a3f6.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6bfe1e209131cea317dd.md>)

[Implementation](<https://github.com/szzexpoi/rex>)

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

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