# Image Classification on Biomedical Domain Average of 10 datasets (test)

Source: [Image Classification on Biomedical Domain Average of 10 datasets (test)](<https://www.sota2.com/research/sota/image-classification-on-biomedical-domain-average-of-10-datasets-test>)

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

Data updated: 2026-04-23T11:03:03.096Z

Average performance of Image Classification models evaluated across a specific collection of 10 biomedical domain datasets, reported on the test split using K-shot accuracy metrics (K=1 to K=16).

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 | Image Classification |
| Dataset (including any stated split) | Biomedical Domain Average of 10 datasets (test) |
| Metric | Accuracy (K=4) |
| Unit | PERCENT |
| Direction | Higher is better |
| Results loaded | 18 of 18 |
| 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 (K=4)](<https://www.sota2.com/research/sota/image-classification-on-biomedical-domain-average-of-10-datasets-test.md?metric=Accuracy%20(K%3D4)>): Higher is better; unit: PERCENT.

- [Accuracy (K=1)](<https://www.sota2.com/research/sota/image-classification-on-biomedical-domain-average-of-10-datasets-test.md?metric=Accuracy%20(K%3D1)>): Higher is better; unit: PERCENT.

- [Accuracy (K=2)](<https://www.sota2.com/research/sota/image-classification-on-biomedical-domain-average-of-10-datasets-test.md?metric=Accuracy%20(K%3D2)>): Higher is better; unit: PERCENT.

- [Accuracy (K=8)](<https://www.sota2.com/research/sota/image-classification-on-biomedical-domain-average-of-10-datasets-test.md?metric=Accuracy%20(K%3D8)>): Higher is better; unit: PERCENT.

- [Accuracy (K=16)](<https://www.sota2.com/research/sota/image-classification-on-biomedical-domain-average-of-10-datasets-test.md?metric=Accuracy%20(K%3D16)>): Higher is better; unit: PERCENT.

## Recorded results

| Method | Accuracy (K=4) | Result date | Report paper |
| --- | --- | --- | --- |
| CLIP-SVD | 68.02 | Not recorded | 68fa7b691e209131cea88420 |
| BiomedCoOp | 64.34 | Not recorded | 68fa7b691e209131cea88420 |
| ProKeR | 63.72 | Not recorded | 68fa7b691e209131cea88420 |
| GDA | 63.41 | Not recorded | 68fa7b691e209131cea88420 |
| DCPL | 62.62 | Not recorded | 68fa7b691e209131cea88420 |
| CLIP-LoRA | 62.31 | Not recorded | 68fa7b691e209131cea88420 |
| Tip-Adapter-F | 62.3 | Not recorded | 68fa7b691e209131cea88420 |
| Linear Probing | 62.12 | Not recorded | 68fa7b691e209131cea88420 |
| ProGrad | 62.1 | Not recorded | 68fa7b691e209131cea88420 |
| CoOp | 61.35 | Not recorded | 68fa7b691e209131cea88420 |
| LP++ | 61.3 | Not recorded | 68fa7b691e209131cea88420 |
| KgCoOp | 60.92 | Not recorded | 68fa7b691e209131cea88420 |
| XCoOp | 60.87 | Not recorded | 68fa7b691e209131cea88420 |
| Tip-Adapter | 58.33 | Not recorded | 68fa7b691e209131cea88420 |
| CoCoOp | 57.63 | Not recorded | 68fa7b691e209131cea88420 |
| CLIP-Adapter | 45.3 | Not recorded | 68fa7b691e209131cea88420 |
| MaPLe | 44.09 | Not recorded | 68fa7b691e209131cea88420 |
| Zero-shot BiomedCLIP | 42.38 | Not recorded | 68fa7b691e209131cea88420 |

### CLIP-SVD

Accuracy (K=4): 68.02. Result ID: 69e9fc4947f34ce924d20ddd.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7b691e209131cea88420.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa7b691e209131cea88420.md>)

Recorded evaluation configuration: {"Evaluation Protocol":"Few-shot adaptation"}. This record does not assert independent reproduction.

### BiomedCoOp

Accuracy (K=4): 64.34. Result ID: 69e9fc4947f34ce924d20ddc.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7b691e209131cea88420.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa77641e209131cea72ffc.md>)

[Implementation](<https://github.com/HealthX-Lab/BiomedCoOp>)

Recorded evaluation configuration: {"Evaluation Protocol":"Few-shot adaptation"}. This record does not assert independent reproduction.

### ProKeR

Accuracy (K=4): 63.72. Result ID: 69e9fc4947f34ce924d20dd7.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7b691e209131cea88420.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa76b01e209131cea6e7a5.md>)

[Implementation](<https://ybendou.github.io/ProKeR/>)

Recorded evaluation configuration: {"Evaluation Protocol":"Few-shot adaptation"}. This record does not assert independent reproduction.

### GDA

Accuracy (K=4): 63.41. Result ID: 69e9fc4947f34ce924d20dd6.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7b691e209131cea88420.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa72141e209131cea545d8.md>)

Recorded evaluation configuration: {"Evaluation Protocol":"Few-shot adaptation"}. This record does not assert independent reproduction.

### DCPL

Accuracy (K=4): 62.62. Result ID: 69e9fc4947f34ce924d20dd9.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7b691e209131cea88420.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa717b1e209131cea5183e.md>)

Recorded evaluation configuration: {"Evaluation Protocol":"Few-shot adaptation"}. This record does not assert independent reproduction.

### CLIP-LoRA

Accuracy (K=4): 62.31. Result ID: 69e9fc4947f34ce924d20dda.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7b691e209131cea88420.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa73741e209131cea5c9ee.md>)

Recorded evaluation configuration: {"Evaluation Protocol":"Few-shot adaptation"}. This record does not assert independent reproduction.

### Tip-Adapter-F

Accuracy (K=4): 62.3. Result ID: 69e9fc4947f34ce924d20dd5.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7b691e209131cea88420.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6b441e209131cea2d1aa.md>)

Recorded evaluation configuration: {"Evaluation Protocol":"Few-shot adaptation"}. This record does not assert independent reproduction.

### Linear Probing

Accuracy (K=4): 62.12. Result ID: 69e9fc4947f34ce924d20dd1.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7b691e209131cea88420.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa69581e209131cea21ad0.md>)

[Implementation](<https://github.com/OpenAI/CLIP>)

Recorded evaluation configuration: {"Evaluation Protocol":"Linear Probing"}. This record does not assert independent reproduction.

### ProGrad

Accuracy (K=4): 62.1. Result ID: 69e9fc4947f34ce924d20dcf.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7b691e209131cea88420.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa7a981e209131cea841f7.md>)

Recorded evaluation configuration: {"Evaluation Protocol":"Few-shot adaptation"}. This record does not assert independent reproduction.

### CoOp

Accuracy (K=4): 61.35. Result ID: 69e9fc4947f34ce924d20dcd.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7b691e209131cea88420.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6d8b1e209131cea3a814.md>)

[Implementation](<https://github.com/KaiyangZhou/CoOp>)

Recorded evaluation configuration: {"Evaluation Protocol":"Few-shot adaptation"}. This record does not assert independent reproduction.

### LP++

Accuracy (K=4): 61.3. Result ID: 69e9fc4947f34ce924d20dd2.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7b691e209131cea88420.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa72bb1e209131cea58378.md>)

Recorded evaluation configuration: {"Evaluation Protocol":"Few-shot adaptation"}. This record does not assert independent reproduction.

### KgCoOp

Accuracy (K=4): 60.92. Result ID: 69e9fc4947f34ce924d20dd0.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7b691e209131cea88420.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6ee91e209131cea42d7a.md>)

Recorded evaluation configuration: {"Evaluation Protocol":"Few-shot adaptation"}. This record does not assert independent reproduction.

### XCoOp

Accuracy (K=4): 60.87. Result ID: 69e9fc4947f34ce924d20dd8.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7b691e209131cea88420.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa727e1e209131cea56ba5.md>)

Recorded evaluation configuration: {"Evaluation Protocol":"Few-shot adaptation"}. This record does not assert independent reproduction.

### Tip-Adapter

Accuracy (K=4): 58.33. Result ID: 69e9fc4947f34ce924d20dd4.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7b691e209131cea88420.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6b441e209131cea2d1aa.md>)

Recorded evaluation configuration: {"Evaluation Protocol":"Few-shot adaptation"}. This record does not assert independent reproduction.

### CoCoOp

Accuracy (K=4): 57.63. Result ID: 69e9fc4947f34ce924d20dce.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7b691e209131cea88420.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6d8c1e209131cea3a831.md>)

[Implementation](<https://github.com/KaiyangZhou/CoOp>)

Recorded evaluation configuration: {"Evaluation Protocol":"Few-shot adaptation"}. This record does not assert independent reproduction.

### CLIP-Adapter

Accuracy (K=4): 45.3. Result ID: 69e9fc4947f34ce924d20dd3.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7b691e209131cea88420.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa77921e209131cea7403b.md>)

Recorded evaluation configuration: {"Evaluation Protocol":"Few-shot adaptation"}. This record does not assert independent reproduction.

### MaPLe

Accuracy (K=4): 44.09. Result ID: 69e9fc4947f34ce924d20ddb.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7b691e209131cea88420.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6ef21e209131cea4311c.md>)

[Implementation](<https://github.com/muzairkhattak/multimodal-prompt-learning>)

Recorded evaluation configuration: {"Evaluation Protocol":"Few-shot adaptation"}. This record does not assert independent reproduction.

### Zero-shot BiomedCLIP

Accuracy (K=4): 42.38. Result ID: 69e9fc4947f34ce924d20dcc.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7b691e209131cea88420.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa768f1e209131cea6db5b.md>)

Recorded evaluation configuration: {"Evaluation Protocol":"Zero-shot"}. This record does not assert independent reproduction.

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