# Medical Visual Question Answering on Med-HallMark

Source: [Medical Visual Question Answering on Med-HallMark](<https://www.sota2.com/research/sota/medical-visual-question-answering-on-med-hallmark>)

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

Data updated: 2026-04-28T09:45:56.166Z

Evaluation of model performance on the Med-HallMark dataset for Medical Visual Question Answering, reporting scores across various NLP and domain-specific 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 | Medical Visual Question Answering |
| Dataset (including any stated split) | Med-HallMark |
| Metric | BertScore |
| Unit | SCALAR |
| Direction | Higher is better |
| Results loaded | 12 of 12 |
| 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

- [BertScore](<https://www.sota2.com/research/sota/medical-visual-question-answering-on-med-hallmark.md?metric=BertScore>): Higher is better; unit: SCALAR.

- [METEOR](<https://www.sota2.com/research/sota/medical-visual-question-answering-on-med-hallmark.md?metric=METEOR>): Higher is better; unit: SCALAR.

- [ROUGE-1](<https://www.sota2.com/research/sota/medical-visual-question-answering-on-med-hallmark.md?metric=ROUGE-1>): Higher is better; unit: SCALAR.

- [ROUGE-2](<https://www.sota2.com/research/sota/medical-visual-question-answering-on-med-hallmark.md?metric=ROUGE-2>): Higher is better; unit: SCALAR.

- [ROUGE-L](<https://www.sota2.com/research/sota/medical-visual-question-answering-on-med-hallmark.md?metric=ROUGE-L>): Higher is better; unit: SCALAR.

- [BLEU](<https://www.sota2.com/research/sota/medical-visual-question-answering-on-med-hallmark.md?metric=BLEU>): Higher is better; unit: SCALAR.

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

- [Accuracy](<https://www.sota2.com/research/sota/medical-visual-question-answering-on-med-hallmark.md?metric=Accuracy>): Higher is better; unit: PERCENT.

## Recorded results

| Method | BertScore | Result date | Report paper |
| --- | --- | --- | --- |
| mPLUG-Owl2 | 55.11 | Not recorded | 68fa73ae1e209131cea5de4d |
| LLaVA1.5-7b | 54.89 | Not recorded | 68fa73ae1e209131cea5de4d |
| LLaVA1.5-13b | 52.82 | Not recorded | 68fa73ae1e209131cea5de4d |
| BLIP2 | 47.97 | Not recorded | 68fa73ae1e209131cea5de4d |
| XrayGPT | 44.4 | Not recorded | 68fa73ae1e209131cea5de4d |
| RadFM | 43.84 | Not recorded | 68fa73ae1e209131cea5de4d |
| mini-gpt4 | 42.93 | Not recorded | 68fa73ae1e209131cea5de4d |
| LLaVA-Med | 36.67 | Not recorded | 68fa73ae1e209131cea5de4d |
| InstructBLIP13b | 36.02 | Not recorded | 68fa73ae1e209131cea5de4d |
| InstructBLIP | 35.99 | Not recorded | 68fa73ae1e209131cea5de4d |
| LLaVA-Med | 35.25 | Not recorded | 68fa73ae1e209131cea5de4d |
| LLaVA-Med | 33.32 | Not recorded | 68fa73ae1e209131cea5de4d |

### mPLUG-Owl2

BertScore: 55.11. Result ID: 69f081c147f34ce924d6db86.

[Reporting paper](<https://www.sota2.com/research/paper/68fa73ae1e209131cea5de4d.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa71201e209131cea4f575.md>)

[Implementation](<https://github.com/X-PLUG/mPLUG-Owl/tree/main/mPLUG-Owl2>)

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

### LLaVA1.5-7b

BertScore: 54.89. Result ID: 69f081c147f34ce924d6db81.

[Reporting paper](<https://www.sota2.com/research/paper/68fa73ae1e209131cea5de4d.md>)

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

### LLaVA1.5-13b

BertScore: 52.82. Result ID: 69f081c147f34ce924d6db82.

[Reporting paper](<https://www.sota2.com/research/paper/68fa73ae1e209131cea5de4d.md>)

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

### BLIP2

BertScore: 47.97. Result ID: 69f081c147f34ce924d6db7e.

[Reporting paper](<https://www.sota2.com/research/paper/68fa73ae1e209131cea5de4d.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6fb71e209131cea46fe6.md>)

[Implementation](<https://github.com/salesforce/LAVIS/tree/main/projects/blip2>)

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

### XrayGPT

BertScore: 44.4. Result ID: 69f081c147f34ce924d6db87.

[Reporting paper](<https://www.sota2.com/research/paper/68fa73ae1e209131cea5de4d.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa78611e209131cea78154.md>)

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

### RadFM

BertScore: 43.84. Result ID: 69f081c147f34ce924d6db89.

[Reporting paper](<https://www.sota2.com/research/paper/68fa73ae1e209131cea5de4d.md>)

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

### mini-gpt4

BertScore: 42.93. Result ID: 69f081c147f34ce924d6db88.

[Reporting paper](<https://www.sota2.com/research/paper/68fa73ae1e209131cea5de4d.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa70bd1e209131cea4cf90.md>)

[Implementation](<https://minigpt-4.github.io/>)

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

### LLaVA-Med

BertScore: 36.67. Result ID: 69f081c147f34ce924d6db83.

[Reporting paper](<https://www.sota2.com/research/paper/68fa73ae1e209131cea5de4d.md>)

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

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

Recorded evaluation configuration: {"Protocol":"Slake-Finetuned"}. This record does not assert independent reproduction.

### InstructBLIP13b

BertScore: 36.02. Result ID: 69f081c147f34ce924d6db80.

[Reporting paper](<https://www.sota2.com/research/paper/68fa73ae1e209131cea5de4d.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6fb71e209131cea47021.md>)

[Implementation](<https://github.com/salesforce/LAVIS/tree/main/projects/instructblip>)

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

### InstructBLIP

BertScore: 35.99. Result ID: 69f081c147f34ce924d6db7f.

[Reporting paper](<https://www.sota2.com/research/paper/68fa73ae1e209131cea5de4d.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6fb71e209131cea47021.md>)

[Implementation](<https://github.com/salesforce/LAVIS/tree/main/projects/instructblip>)

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

### LLaVA-Med

BertScore: 35.25. Result ID: 69f081c147f34ce924d6db84.

[Reporting paper](<https://www.sota2.com/research/paper/68fa73ae1e209131cea5de4d.md>)

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

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

Recorded evaluation configuration: {"Protocol":"Rad-Finetuned"}. This record does not assert independent reproduction.

### LLaVA-Med

BertScore: 33.32. Result ID: 69f081c147f34ce924d6db85.

[Reporting paper](<https://www.sota2.com/research/paper/68fa73ae1e209131cea5de4d.md>)

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

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

Recorded evaluation configuration: {"Protocol":"Pathvqa-Finetuned"}. This record does not assert independent reproduction.

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