# Image Classification on HAM (test)

Source: [Image Classification on HAM (test)](<https://www.sota2.com/research/sota/image-classification-on-ham-test>)

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

Data updated: 2026-02-26T05:31:34.933Z

Evaluation of model performance on the test split of the HAM dataset for general Image Classification.

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) | HAM (test) |
| Metric | Accuracy |
| Unit | PERCENT |
| Direction | Higher is better |
| Results loaded | 9 of 9 |
| 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/image-classification-on-ham-test.md?metric=Accuracy>): Higher is better; unit: PERCENT.

## Recorded results

| Method | Accuracy | Result date | Report paper |
| --- | --- | --- | --- |
| Linear Probe | 83.18 | Not recorded | 69705d6ae25ad2fa9d1ecf25 |
| LaBo | 81.39 | Not recorded | 69705d6ae25ad2fa9d1ecf25 |
| PCBM-ReD | 81.39 | Not recorded | 69705d6ae25ad2fa9d1ecf25 |
| V2C-CBM | 81.12 | Not recorded | 69705d6ae25ad2fa9d1ecf25 |
| Res-CBM | 75.72 | Not recorded | 69705d6ae25ad2fa9d1ecf25 |
| PCBM-ReD | 58.51 | Not recorded | 69705d6ae25ad2fa9d1ecf25 |
| CLIP | 58.21 | Not recorded | 69705d6ae25ad2fa9d1ecf25 |
| PCBM-ReD + CuPL | 58.11 | Not recorded | 69705d6ae25ad2fa9d1ecf25 |
| CuPL | 57.91 | Not recorded | 69705d6ae25ad2fa9d1ecf25 |

### Linear Probe

Accuracy: 83.18. Result ID: 6997afed5698665ec4061b3d.

[Reporting paper](<https://www.sota2.com/research/paper/69705d6ae25ad2fa9d1ecf25.md>)

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

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

Recorded evaluation configuration: {"Interpretability":false,"Evaluation Setting":"Fully-supervised","Backbone":"CLIP ViT-L/14"}. This record does not assert independent reproduction.

### LaBo

Accuracy: 81.39. Result ID: 6997afed5698665ec4061b3e.

[Reporting paper](<https://www.sota2.com/research/paper/69705d6ae25ad2fa9d1ecf25.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6f231e209131cea43f60.md>)

Recorded evaluation configuration: {"Interpretability":true,"Evaluation Setting":"Fully-supervised","Backbone":"CLIP ViT-L/14"}. This record does not assert independent reproduction.

### PCBM-ReD

Accuracy: 81.39. Result ID: 6997afed5698665ec4061b41.

[Reporting paper](<https://www.sota2.com/research/paper/69705d6ae25ad2fa9d1ecf25.md>)

[Method origin paper](<https://www.sota2.com/research/paper/69705d6ae25ad2fa9d1ecf25.md>)

Recorded evaluation configuration: {"Interpretability":true,"Evaluation Setting":"Fully-supervised","Backbone":"CLIP ViT-L/14"}. This record does not assert independent reproduction.

### V2C-CBM

Accuracy: 81.12. Result ID: 6997afed5698665ec4061b40.

[Reporting paper](<https://www.sota2.com/research/paper/69705d6ae25ad2fa9d1ecf25.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa79941e209131cea7e642.md>)

Recorded evaluation configuration: {"Interpretability":true,"Evaluation Setting":"Fully-supervised","Backbone":"CLIP ViT-L/14"}. This record does not assert independent reproduction.

### Res-CBM

Accuracy: 75.72. Result ID: 6997afed5698665ec4061b3f.

[Reporting paper](<https://www.sota2.com/research/paper/69705d6ae25ad2fa9d1ecf25.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa72f41e209131cea598ce.md>)

[Implementation](<https://github.com/HelloSCM/Res-CBM>)

Recorded evaluation configuration: {"Interpretability":true,"Evaluation Setting":"Fully-supervised","Backbone":"CLIP ViT-L/14"}. This record does not assert independent reproduction.

### PCBM-ReD

Accuracy: 58.51. Result ID: 6997afed5698665ec4061b43.

[Reporting paper](<https://www.sota2.com/research/paper/69705d6ae25ad2fa9d1ecf25.md>)

[Method origin paper](<https://www.sota2.com/research/paper/69705d6ae25ad2fa9d1ecf25.md>)

Recorded evaluation configuration: {"Interpretability":true,"Evaluation Setting":"Zero-shot","Backbone":"CLIP ViT-L/14"}. This record does not assert independent reproduction.

### CLIP

Accuracy: 58.21. Result ID: 6997afed5698665ec4061b42.

[Reporting paper](<https://www.sota2.com/research/paper/69705d6ae25ad2fa9d1ecf25.md>)

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

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

Recorded evaluation configuration: {"Interpretability":false,"Evaluation Setting":"Zero-shot","Backbone":"CLIP ViT-L/14"}. This record does not assert independent reproduction.

### PCBM-ReD + CuPL

Accuracy: 58.11. Result ID: 6997afed5698665ec4061b45.

[Reporting paper](<https://www.sota2.com/research/paper/69705d6ae25ad2fa9d1ecf25.md>)

[Method origin paper](<https://www.sota2.com/research/paper/69705d6ae25ad2fa9d1ecf25.md>)

Recorded evaluation configuration: {"Interpretability":true,"Evaluation Setting":"Zero-shot","Backbone":"CLIP ViT-L/14"}. This record does not assert independent reproduction.

### CuPL

Accuracy: 57.91. Result ID: 6997afed5698665ec4061b44.

[Reporting paper](<https://www.sota2.com/research/paper/69705d6ae25ad2fa9d1ecf25.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa71551e209131cea508eb.md>)

Recorded evaluation configuration: {"Interpretability":false,"Evaluation Setting":"Zero-shot","Backbone":"CLIP ViT-L/14"}. This record does not assert independent reproduction.

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