# Multi-label Medical Image Classification on ChestXray14

Source: [Multi-label Medical Image Classification on ChestXray14](<https://www.sota2.com/research/sota/multi-label-medical-image-classification-on-chestxray14>)

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

Data updated: 2026-04-23T07:35:03.681Z

Evaluation of multi-label medical image classification performance on the ChestXray14 dataset, focusing on robustness metrics including AUC under clean conditions and PGD adversarial attacks.

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 | Multi-label Medical Image Classification |
| Dataset (including any stated split) | ChestXray14 |
| Metric | AUC (Clean) |
| 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

- [AUC (Clean)](<https://www.sota2.com/research/sota/multi-label-medical-image-classification-on-chestxray14.md?metric=AUC%20(Clean)>): Higher is better; unit: PERCENT.

- [AUC (PGD Attack)](<https://www.sota2.com/research/sota/multi-label-medical-image-classification-on-chestxray14.md?metric=AUC%20(PGD%20Attack)>): Higher is better; unit: PERCENT.

## Recorded results

| Method | AUC (Clean) | Result date | Report paper |
| --- | --- | --- | --- |
| HITA (2 trees) | 73 | Not recorded | 69e85bd47d0172140ec81123 |
| HITA | 72.7 | Not recorded | 69e85bd47d0172140ec81123 |
| AoS | 71.8 | Not recorded | 69e85bd47d0172140ec81123 |
| PMG-FT | 69.2 | Not recorded | 69e85bd47d0172140ec81123 |
| FARE | 68.7 | Not recorded | 69e85bd47d0172140ec81123 |
| TeCoA | 67.4 | Not recorded | 69e85bd47d0172140ec81123 |

### HITA (2 trees)

AUC (Clean): 73. Result ID: 69e9cb8a47f34ce924d09d79.

[Reporting paper](<https://www.sota2.com/research/paper/69e85bd47d0172140ec81123.md>)

[Method origin paper](<https://www.sota2.com/research/paper/69e85bd47d0172140ec81123.md>)

Recorded evaluation configuration: {"Evaluation Protocol":"Zero-shot","Adversarial Attack":"PGD-20"}. This record does not assert independent reproduction.

### HITA

AUC (Clean): 72.7. Result ID: 69e9cb8a47f34ce924d09d78.

[Reporting paper](<https://www.sota2.com/research/paper/69e85bd47d0172140ec81123.md>)

[Method origin paper](<https://www.sota2.com/research/paper/69e85bd47d0172140ec81123.md>)

Recorded evaluation configuration: {"Evaluation Protocol":"Zero-shot","Adversarial Attack":"PGD-20"}. This record does not assert independent reproduction.

### AoS

AUC (Clean): 71.8. Result ID: 69e9cb8a47f34ce924d09d77.

[Reporting paper](<https://www.sota2.com/research/paper/69e85bd47d0172140ec81123.md>)

Recorded evaluation configuration: {"Evaluation Protocol":"Zero-shot","Adversarial Attack":"PGD-20"}. This record does not assert independent reproduction.

### PMG-FT

AUC (Clean): 69.2. Result ID: 69e9cb8a47f34ce924d09d75.

[Reporting paper](<https://www.sota2.com/research/paper/69e85bd47d0172140ec81123.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa72df1e209131cea5912e.md>)

Recorded evaluation configuration: {"Evaluation Protocol":"Zero-shot","Adversarial Attack":"PGD-20"}. This record does not assert independent reproduction.

### FARE

AUC (Clean): 68.7. Result ID: 69e9cb8a47f34ce924d09d76.

[Reporting paper](<https://www.sota2.com/research/paper/69e85bd47d0172140ec81123.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa738c1e209131cea5d24a.md>)

Recorded evaluation configuration: {"Evaluation Protocol":"Zero-shot","Adversarial Attack":"PGD-20"}. This record does not assert independent reproduction.

### TeCoA

AUC (Clean): 67.4. Result ID: 69e9cb8a47f34ce924d09d74.

[Reporting paper](<https://www.sota2.com/research/paper/69e85bd47d0172140ec81123.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6f271e209131cea44177.md>)

Recorded evaluation configuration: {"Evaluation Protocol":"Zero-shot","Adversarial Attack":"PGD-20"}. This record does not assert independent reproduction.

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