# ROI-level Visual Question Answering on SMU-281 GI & HPB

Source: [ROI-level Visual Question Answering on SMU-281 GI & HPB](<https://www.sota2.com/research/sota/roi-level-visual-question-answering-on-smu-281-gi-and-hpb>)

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

Data updated: 2026-06-09T08:03:21.982Z

Evaluation of model performance on the SMU-281 GI & HPB dataset specifically for ROI-level Visual Question Answering tasks.

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 | ROI-level Visual Question Answering |
| Dataset (including any stated split) | SMU-281 GI & HPB |
| Metric | Accuracy |
| Unit | PERCENT |
| Direction | Higher is better |
| Results loaded | 6 of 6 |
| 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/roi-level-visual-question-answering-on-smu-281-gi-and-hpb.md?metric=Accuracy>): Higher is better; unit: PERCENT.

## Recorded results

| Method | Accuracy | Result date | Report paper |
| --- | --- | --- | --- |
| PathPocket | 69.4 | Not recorded | 6a279e3f90966ade919bf862 |
| Qwen3VL-8B | 52.8 | Not recorded | 6a279e3f90966ade919bf862 |
| Qwen3VL-2B | 47.2 | Not recorded | 6a279e3f90966ade919bf862 |
| Qwen3VL-4B | 47.2 | Not recorded | 6a279e3f90966ade919bf862 |
| Quilt-LLaVA | 19.4 | Not recorded | 6a279e3f90966ade919bf862 |
| LLaVA-Med | 8.3 | Not recorded | 6a279e3f90966ade919bf862 |

### PathPocket

Accuracy: 69.4. Result ID: 6a27c882fbc373592c245e08.

[Reporting paper](<https://www.sota2.com/research/paper/6a279e3f90966ade919bf862.md>)

[Method origin paper](<https://www.sota2.com/research/paper/6a279e3f90966ade919bf862.md>)

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

### Qwen3VL-8B

Accuracy: 52.8. Result ID: 6a27c882fbc373592c245e05.

[Reporting paper](<https://www.sota2.com/research/paper/6a279e3f90966ade919bf862.md>)

[Method origin paper](<https://www.sota2.com/research/paper/6927db54d6b0dd3342930f66.md>)

Recorded evaluation configuration: {"Model size":"8B"}. This record does not assert independent reproduction.

### Qwen3VL-2B

Accuracy: 47.2. Result ID: 6a27c882fbc373592c245e03.

[Reporting paper](<https://www.sota2.com/research/paper/6a279e3f90966ade919bf862.md>)

[Method origin paper](<https://www.sota2.com/research/paper/6927db54d6b0dd3342930f66.md>)

Recorded evaluation configuration: {"Model size":"2B"}. This record does not assert independent reproduction.

### Qwen3VL-4B

Accuracy: 47.2. Result ID: 6a27c882fbc373592c245e04.

[Reporting paper](<https://www.sota2.com/research/paper/6a279e3f90966ade919bf862.md>)

[Method origin paper](<https://www.sota2.com/research/paper/6927db54d6b0dd3342930f66.md>)

Recorded evaluation configuration: {"Model size":"4B"}. This record does not assert independent reproduction.

### Quilt-LLaVA

Accuracy: 19.4. Result ID: 6a27c882fbc373592c245e07.

[Reporting paper](<https://www.sota2.com/research/paper/6a279e3f90966ade919bf862.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa76881e209131cea6d7e8.md>)

[Implementation](<https://www.sota2.com/quilt-llava.github.io>)

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

### LLaVA-Med

Accuracy: 8.3. Result ID: 6a27c882fbc373592c245e06.

[Reporting paper](<https://www.sota2.com/research/paper/6a279e3f90966ade919bf862.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6f8e1e209131cea46356.md>)

[Implementation](<https://aka.ms/llava-med>)

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.