# Medical Visual Question Answering on SLAKE Closed (AUROC)

Source: [Medical Visual Question Answering on SLAKE Closed (AUROC)](<https://www.sota2.com/research/sota/medical-visual-question-answering-on-slake-closed-auroc>)

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

Data updated: 2026-04-06T07:47:09.953Z

Evaluation of model performance on the closed split of the SLAKE dataset for Medical Visual Question Answering, reporting AUROC.

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) | SLAKE Closed |
| Metric | AUROC |
| Unit | PERCENT |
| Direction | Higher is better |
| Results loaded | 17 of 17 |
| 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

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

## Recorded results

| Method | AUROC | Result date | Report paper |
| --- | --- | --- | --- |
| HAC-Platt | 73.2 | Not recorded | 69d33e177d0172140ec7377c |
| HAC-Gate | 70 | Not recorded | 69d33e177d0172140ec7377c |
| HAC-Platt | 69 | Not recorded | 69d33e177d0172140ec7377c |
| HAC-Platt | 68.6 | Not recorded | 69d33e177d0172140ec7377c |
| HAC-Gate | 68.6 | Not recorded | 69d33e177d0172140ec7377c |
| HAC-Platt | 65.7 | Not recorded | 69d33e177d0172140ec7377c |
| HAC-Platt | 65.6 | Not recorded | 69d33e177d0172140ec7377c |
| HAC-Gate | 65.6 | Not recorded | 69d33e177d0172140ec7377c |
| HAC-Gate | 65.4 | Not recorded | 69d33e177d0172140ec7377c |
| Uncalibrated | 60.7 | Not recorded | 69d33e177d0172140ec7377c |
| Platt Scaling | 60.7 | Not recorded | 69d33e177d0172140ec7377c |
| Isotonic Regr. | 60.7 | Not recorded | 69d33e177d0172140ec7377c |
| Uncalibrated | 58.6 | Not recorded | 69d33e177d0172140ec7377c |
| Platt Scaling | 58.6 | Not recorded | 69d33e177d0172140ec7377c |
| Isotonic Regr. | 58.6 | Not recorded | 69d33e177d0172140ec7377c |
| Hist. Binning | 56.6 | Not recorded | 69d33e177d0172140ec7377c |
| Hist. Binning | 50 | Not recorded | 69d33e177d0172140ec7377c |

### HAC-Platt

AUROC: 73.2. Result ID: 69d364de47f34ce924baf4b5.

[Reporting paper](<https://www.sota2.com/research/paper/69d33e177d0172140ec7377c.md>)

[Method origin paper](<https://www.sota2.com/research/paper/69d33e177d0172140ec7377c.md>)

Recorded evaluation configuration: {"Model":"InternVL3-2B","Confidence":"Samp."}. This record does not assert independent reproduction.

### HAC-Gate

AUROC: 70. Result ID: 69d364de47f34ce924baf4b6.

[Reporting paper](<https://www.sota2.com/research/paper/69d33e177d0172140ec7377c.md>)

[Method origin paper](<https://www.sota2.com/research/paper/69d33e177d0172140ec7377c.md>)

Recorded evaluation configuration: {"Model":"InternVL3-38B","Confidence":"Samp."}. This record does not assert independent reproduction.

### HAC-Platt

AUROC: 69. Result ID: 69d364de47f34ce924baf4b4.

[Reporting paper](<https://www.sota2.com/research/paper/69d33e177d0172140ec7377c.md>)

[Method origin paper](<https://www.sota2.com/research/paper/69d33e177d0172140ec7377c.md>)

Recorded evaluation configuration: {"Model":"Qwen3-VL-32B","Confidence":"Verb."}. This record does not assert independent reproduction.

### HAC-Platt

AUROC: 68.6. Result ID: 69d364de47f34ce924baf4b2.

[Reporting paper](<https://www.sota2.com/research/paper/69d33e177d0172140ec7377c.md>)

[Method origin paper](<https://www.sota2.com/research/paper/69d33e177d0172140ec7377c.md>)

Recorded evaluation configuration: {"Model":"Qwen3-VL-8B","Confidence":"Samp."}. This record does not assert independent reproduction.

### HAC-Gate

AUROC: 68.6. Result ID: 69d364de47f34ce924baf4b3.

[Reporting paper](<https://www.sota2.com/research/paper/69d33e177d0172140ec7377c.md>)

[Method origin paper](<https://www.sota2.com/research/paper/69d33e177d0172140ec7377c.md>)

Recorded evaluation configuration: {"Model":"Qwen3-VL-8B","Confidence":"Samp."}. This record does not assert independent reproduction.

### HAC-Platt

AUROC: 65.7. Result ID: 69d364de47f34ce924baf4af.

[Reporting paper](<https://www.sota2.com/research/paper/69d33e177d0172140ec7377c.md>)

[Method origin paper](<https://www.sota2.com/research/paper/69d33e177d0172140ec7377c.md>)

Recorded evaluation configuration: {"Model":"Qwen3-VL-2B","Confidence":"Verb."}. This record does not assert independent reproduction.

### HAC-Platt

AUROC: 65.6. Result ID: 69d364de47f34ce924baf4ae.

[Reporting paper](<https://www.sota2.com/research/paper/69d33e177d0172140ec7377c.md>)

[Method origin paper](<https://www.sota2.com/research/paper/69d33e177d0172140ec7377c.md>)

Recorded evaluation configuration: {"Model":"Qwen3-VL-2B","Confidence":"Samp."}. This record does not assert independent reproduction.

### HAC-Gate

AUROC: 65.6. Result ID: 69d364de47f34ce924baf4b0.

[Reporting paper](<https://www.sota2.com/research/paper/69d33e177d0172140ec7377c.md>)

[Method origin paper](<https://www.sota2.com/research/paper/69d33e177d0172140ec7377c.md>)

Recorded evaluation configuration: {"Model":"Qwen3-VL-2B","Confidence":"Samp."}. This record does not assert independent reproduction.

### HAC-Gate

AUROC: 65.4. Result ID: 69d364de47f34ce924baf4b1.

[Reporting paper](<https://www.sota2.com/research/paper/69d33e177d0172140ec7377c.md>)

[Method origin paper](<https://www.sota2.com/research/paper/69d33e177d0172140ec7377c.md>)

Recorded evaluation configuration: {"Model":"Qwen3-VL-2B","Confidence":"Verb."}. This record does not assert independent reproduction.

### Uncalibrated

AUROC: 60.7. Result ID: 69d364de47f34ce924baf4a6.

[Reporting paper](<https://www.sota2.com/research/paper/69d33e177d0172140ec7377c.md>)

Recorded evaluation configuration: {"Model":"Qwen3-VL-2B","Confidence":"Samp."}. This record does not assert independent reproduction.

### Platt Scaling

AUROC: 60.7. Result ID: 69d364de47f34ce924baf4a8.

[Reporting paper](<https://www.sota2.com/research/paper/69d33e177d0172140ec7377c.md>)

Recorded evaluation configuration: {"Model":"Qwen3-VL-2B","Confidence":"Samp."}. This record does not assert independent reproduction.

### Isotonic Regr.

AUROC: 60.7. Result ID: 69d364de47f34ce924baf4aa.

[Reporting paper](<https://www.sota2.com/research/paper/69d33e177d0172140ec7377c.md>)

Recorded evaluation configuration: {"Model":"Qwen3-VL-2B","Confidence":"Samp."}. This record does not assert independent reproduction.

### Uncalibrated

AUROC: 58.6. Result ID: 69d364de47f34ce924baf4a7.

[Reporting paper](<https://www.sota2.com/research/paper/69d33e177d0172140ec7377c.md>)

Recorded evaluation configuration: {"Model":"Qwen3-VL-2B","Confidence":"Verb."}. This record does not assert independent reproduction.

### Platt Scaling

AUROC: 58.6. Result ID: 69d364de47f34ce924baf4a9.

[Reporting paper](<https://www.sota2.com/research/paper/69d33e177d0172140ec7377c.md>)

Recorded evaluation configuration: {"Model":"Qwen3-VL-2B","Confidence":"Verb."}. This record does not assert independent reproduction.

### Isotonic Regr.

AUROC: 58.6. Result ID: 69d364de47f34ce924baf4ab.

[Reporting paper](<https://www.sota2.com/research/paper/69d33e177d0172140ec7377c.md>)

Recorded evaluation configuration: {"Model":"Qwen3-VL-2B","Confidence":"Verb."}. This record does not assert independent reproduction.

### Hist. Binning

AUROC: 56.6. Result ID: 69d364de47f34ce924baf4ac.

[Reporting paper](<https://www.sota2.com/research/paper/69d33e177d0172140ec7377c.md>)

Recorded evaluation configuration: {"Model":"Qwen3-VL-2B","Confidence":"Samp."}. This record does not assert independent reproduction.

### Hist. Binning

AUROC: 50. Result ID: 69d364de47f34ce924baf4ad.

[Reporting paper](<https://www.sota2.com/research/paper/69d33e177d0172140ec7377c.md>)

Recorded evaluation configuration: {"Model":"Qwen3-VL-2B","Confidence":"Verb."}. This record does not assert independent reproduction.

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