# Visual Question Answering on TextVQA (Accuracy Single/Multi)

Source: [Visual Question Answering on TextVQA (Accuracy Single/Multi)](<https://www.sota2.com/research/sota/visual-question-answering-on-textvqa-accuracy-single-multi>)

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

Data updated: 2026-02-26T05:34:26.995Z

Evaluation of model performance on the TextVQA dataset for Visual Question Answering, focusing specifically on Single and Multi-answer accuracy 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 |
| Metric | Accuracy (Single) |
| Unit | PERCENT |
| Direction | Higher is better |
| Results loaded | 7 of 7 |
| 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 (Single)](<https://www.sota2.com/research/sota/visual-question-answering-on-textvqa-accuracy-single-multi.md?metric=Accuracy%20(Single)>): Higher is better; unit: PERCENT.

- [Accuracy (Multi)](<https://www.sota2.com/research/sota/visual-question-answering-on-textvqa-accuracy-single-multi.md?metric=Accuracy%20(Multi)>): Higher is better; unit: PERCENT.

## Recorded results

| Method | Accuracy (Single) | Result date | Report paper |
| --- | --- | --- | --- |
| CoCoT (Qwen) | 75.9 | Not recorded | 695de85d53fdd5c361743977 |
| VisCoT (Qwen) | 71.9 | Not recorded | 695de85d53fdd5c361743977 |
| Qwen2-VL | 71.3 | Not recorded | 695de85d53fdd5c361743977 |
| CoCoT\* (LLaVA) | 53.3 | Not recorded | 695de85d53fdd5c361743977 |
| VisCoT\* (LLaVA) | 51.9 | Not recorded | 695de85d53fdd5c361743977 |
| CoCoT (LLaVA) | 47.6 | Not recorded | 695de85d53fdd5c361743977 |
| LLaVA-1.5 | 13.5 | Not recorded | 695de85d53fdd5c361743977 |

### CoCoT (Qwen)

Accuracy (Single): 75.9. Result ID: 6997ee545698665ec4478a49.

[Reporting paper](<https://www.sota2.com/research/paper/695de85d53fdd5c361743977.md>)

[Method origin paper](<https://www.sota2.com/research/paper/695de85d53fdd5c361743977.md>)

Recorded evaluation configuration: {"Backbone":"Qwen2-VL 7B","Training samples":"0"}. This record does not assert independent reproduction.

### VisCoT (Qwen)

Accuracy (Single): 71.9. Result ID: 6997ee545698665ec4478a4a.

[Reporting paper](<https://www.sota2.com/research/paper/695de85d53fdd5c361743977.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa758f1e209131cea68a1f.md>)

Recorded evaluation configuration: {"Backbone":"Qwen2-VL 7B","Training samples":"0"}. This record does not assert independent reproduction.

### Qwen2-VL

Accuracy (Single): 71.3. Result ID: 6997ee545698665ec4478a48.

[Reporting paper](<https://www.sota2.com/research/paper/695de85d53fdd5c361743977.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa75151e209131cea65976.md>)

[Implementation](<https://github.com/QwenLM/Qwen2-VL>)

Recorded evaluation configuration: {"Backbone":"Qwen2-VL 7B","Training samples":"0"}. This record does not assert independent reproduction.

### CoCoT\* (LLaVA)

Accuracy (Single): 53.3. Result ID: 6997ee545698665ec4478a47.

[Reporting paper](<https://www.sota2.com/research/paper/695de85d53fdd5c361743977.md>)

[Method origin paper](<https://www.sota2.com/research/paper/695de85d53fdd5c361743977.md>)

Recorded evaluation configuration: {"Backbone":"LLaVA-1.5 7B","Training samples":"14k","Trained on annotated data":true}. This record does not assert independent reproduction.

### VisCoT\* (LLaVA)

Accuracy (Single): 51.9. Result ID: 6997ee545698665ec4478a46.

[Reporting paper](<https://www.sota2.com/research/paper/695de85d53fdd5c361743977.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa758f1e209131cea68a1f.md>)

Recorded evaluation configuration: {"Backbone":"LLaVA-1.5 7B","Training samples":"363k","Trained on annotated data":true}. This record does not assert independent reproduction.

### CoCoT (LLaVA)

Accuracy (Single): 47.6. Result ID: 6997ee545698665ec4478a45.

[Reporting paper](<https://www.sota2.com/research/paper/695de85d53fdd5c361743977.md>)

[Method origin paper](<https://www.sota2.com/research/paper/695de85d53fdd5c361743977.md>)

Recorded evaluation configuration: {"Backbone":"LLaVA-1.5 7B","Training samples":"0"}. This record does not assert independent reproduction.

### LLaVA-1.5

Accuracy (Single): 13.5. Result ID: 6997ee545698665ec4478a44.

[Reporting paper](<https://www.sota2.com/research/paper/695de85d53fdd5c361743977.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa732e1e209131cea5af59.md>)

[Implementation](<https://llava-vl.github.io/>)

Recorded evaluation configuration: {"Backbone":"LLaVA-1.5 7B","Training samples":"0"}. This record does not assert independent reproduction.

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