# Visual Question Answering on TextVQA OCR n=5000 (val)

Source: [Visual Question Answering on TextVQA OCR n=5000 (val)](<https://www.sota2.com/research/sota/visual-question-answering-on-textvqa-ocr-n-5000-val>)

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

Data updated: 2026-05-08T11:02:58.223Z

Evaluation of Visual Question Answering performance on the specific 5000-sample validation subset of the TextVQA dataset, reporting answer loss and delta metrics.

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) | TextVQA OCR n=5000 (val) |
| Metric | Answer Loss |
| Unit | SCALAR |
| Direction | Lower 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

- [Answer Loss](<https://www.sota2.com/research/sota/visual-question-answering-on-textvqa-ocr-n-5000-val.md?metric=Answer%20Loss>): Lower is better; unit: SCALAR.

- [Delta vs Π3](<https://www.sota2.com/research/sota/visual-question-answering-on-textvqa-ocr-n-5000-val.md?metric=Delta%20vs%20%CE%A03>): Higher is better; unit: SCALAR.

## Recorded results

| Method | Answer Loss | Result date | Report paper |
| --- | --- | --- | --- |
| Π3 | 0.8212 | Not recorded | 69fd6e565497b3b670c896a1 |
| Π1 | 0.8445 | Not recorded | 69fd6e565497b3b670c896a1 |
| Π0 | 0.8463 | Not recorded | 69fd6e565497b3b670c896a1 |
| Π2 | 0.8595 | Not recorded | 69fd6e565497b3b670c896a1 |

### Π3

Answer Loss: 0.8212. Result ID: 69fdc2d647f34ce924e18fd8.

[Reporting paper](<https://www.sota2.com/research/paper/69fd6e565497b3b670c896a1.md>)

[Method origin paper](<https://www.sota2.com/research/paper/69fd6e565497b3b670c896a1.md>)

Recorded evaluation configuration: {"best family":"OCR gate + fallback"}. This record does not assert independent reproduction.

### Π1

Answer Loss: 0.8445. Result ID: 69fdc2d647f34ce924e18fd6.

[Reporting paper](<https://www.sota2.com/research/paper/69fd6e565497b3b670c896a1.md>)

[Method origin paper](<https://www.sota2.com/research/paper/69fd6e565497b3b670c896a1.md>)

Recorded evaluation configuration: {"best family":"CART-d4"}. This record does not assert independent reproduction.

### Π0

Answer Loss: 0.8463. Result ID: 69fdc2d647f34ce924e18fd5.

[Reporting paper](<https://www.sota2.com/research/paper/69fd6e565497b3b670c896a1.md>)

[Method origin paper](<https://www.sota2.com/research/paper/69fd6e565497b3b670c896a1.md>)

Recorded evaluation configuration: {"best family":"always\_ocr"}. This record does not assert independent reproduction.

### Π2

Answer Loss: 0.8595. Result ID: 69fdc2d647f34ce924e18fd7.

[Reporting paper](<https://www.sota2.com/research/paper/69fd6e565497b3b670c896a1.md>)

[Method origin paper](<https://www.sota2.com/research/paper/69fd6e565497b3b670c896a1.md>)

Recorded evaluation configuration: {"best family":"logistic C=0.3"}. This record does not assert independent reproduction.

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