# Classification on CheXpert 5x200 1.0

Source: [Classification on CheXpert 5x200 1.0](<https://www.sota2.com/research/sota/classification-on-chexpert-5x200-1-0>)

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

Data updated: 2026-02-26T05:29:54.323Z

Evaluation of general classification performance on the specific CheXpert 5x200 version 1.0 subset, reporting Accuracy and AUC.

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 | Classification |
| Dataset (including any stated split) | CheXpert 5x200 1.0 |
| Metric | Accuracy |
| Unit | PERCENT |
| Direction | Higher is better |
| Results loaded | 27 of 27 |
| 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/classification-on-chexpert-5x200-1-0.md?metric=Accuracy>): Higher is better; unit: PERCENT.

- [AUC](<https://www.sota2.com/research/sota/classification-on-chexpert-5x200-1-0.md?metric=AUC>): Higher is better; unit: PERCENT.

## Recorded results

| Method | Accuracy | Result date | Report paper |
| --- | --- | --- | --- |
| CLEFT-LORA | 66.74 | Not recorded | 68fa74541e209131cea611b9 |
| CLEFT-LORA | 63.96 | Not recorded | 68fa74541e209131cea611b9 |
| CLEFT-LORA | 63.46 | Not recorded | 68fa74541e209131cea611b9 |
| CLEFT-Prefix | 61.56 | Not recorded | 68fa74541e209131cea611b9 |
| CLEFT-IA3 | 61.06 | Not recorded | 68fa74541e209131cea611b9 |
| CLEFT-Prefix | 60.55 | Not recorded | 68fa74541e209131cea611b9 |
| CLEFT-IA3 | 60.16 | Not recorded | 68fa74541e209131cea611b9 |
| MedCLIP-Swin | 59.42 | Not recorded | 68fa74541e209131cea611b9 |
| MRM-VIT | 58.26 | Not recorded | 68fa74541e209131cea611b9 |
| GLORIA-R50 | 57.56 | Not recorded | 68fa74541e209131cea611b9 |
| MedCLIP-Swin | 57.46 | Not recorded | 68fa74541e209131cea611b9 |
| MGCA-VIT+ | 56.96 | Not recorded | 68fa74541e209131cea611b9 |
| MRM-VIT | 56.56 | Not recorded | 68fa74541e209131cea611b9 |
| ImageNet-ViT | 56.46 | Not recorded | 68fa74541e209131cea611b9 |
| MedCLIP-Swin | 54.55 | Not recorded | 68fa74541e209131cea611b9 |
| GLORIA-R50 | 51.65 | Not recorded | 68fa74541e209131cea611b9 |
| MGCA-VIT+ | 51.35 | Not recorded | 68fa74541e209131cea611b9 |
| ConVIRT-R50 | 47.7 | Not recorded | 68fa74541e209131cea611b9 |
| CLIP-VIT | 44.84 | Not recorded | 68fa74541e209131cea611b9 |
| GLORIA-R50 | 43.28 | Not recorded | 68fa74541e209131cea611b9 |
| ConVIRT-R50 | 41.88 | Not recorded | 68fa74541e209131cea611b9 |
| ImageNet-ViT | 38.54 | Not recorded | 68fa74541e209131cea611b9 |
| CLIP-VIT | 35.34 | Not recorded | 68fa74541e209131cea611b9 |
| MGCA-VIT+ | 26.63 | Not recorded | 68fa74541e209131cea611b9 |
| Random-ViT | 20.62 | Not recorded | 68fa74541e209131cea611b9 |
| Random-ViT | 20.32 | Not recorded | 68fa74541e209131cea611b9 |
| CLIP-VIT | 20.16 | Not recorded | 68fa74541e209131cea611b9 |

### CLEFT-LORA

Accuracy: 66.74. Result ID: 6993501d3a6c314849151c60.

[Reporting paper](<https://www.sota2.com/research/paper/68fa74541e209131cea611b9.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa74541e209131cea611b9.md>)

[Implementation](<https://github.com/XYPB/CLEFT>)

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

### CLEFT-LORA

Accuracy: 63.96. Result ID: 6993501d3a6c314849151c62.

[Reporting paper](<https://www.sota2.com/research/paper/68fa74541e209131cea611b9.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa74541e209131cea611b9.md>)

[Implementation](<https://github.com/XYPB/CLEFT>)

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

### CLEFT-LORA

Accuracy: 63.46. Result ID: 6993501d3a6c314849151c61.

[Reporting paper](<https://www.sota2.com/research/paper/68fa74541e209131cea611b9.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa74541e209131cea611b9.md>)

[Implementation](<https://github.com/XYPB/CLEFT>)

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

### CLEFT-Prefix

Accuracy: 61.56. Result ID: 6993501d3a6c314849151c5d.

[Reporting paper](<https://www.sota2.com/research/paper/68fa74541e209131cea611b9.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa74541e209131cea611b9.md>)

[Implementation](<https://github.com/XYPB/CLEFT>)

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

### CLEFT-IA3

Accuracy: 61.06. Result ID: 6993501d3a6c314849151c5f.

[Reporting paper](<https://www.sota2.com/research/paper/68fa74541e209131cea611b9.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa74541e209131cea611b9.md>)

[Implementation](<https://github.com/XYPB/CLEFT>)

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

### CLEFT-Prefix

Accuracy: 60.55. Result ID: 6993501d3a6c314849151c5c.

[Reporting paper](<https://www.sota2.com/research/paper/68fa74541e209131cea611b9.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa74541e209131cea611b9.md>)

[Implementation](<https://github.com/XYPB/CLEFT>)

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

### CLEFT-IA3

Accuracy: 60.16. Result ID: 6993501d3a6c314849151c5e.

[Reporting paper](<https://www.sota2.com/research/paper/68fa74541e209131cea611b9.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa74541e209131cea611b9.md>)

[Implementation](<https://github.com/XYPB/CLEFT>)

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

### MedCLIP-Swin

Accuracy: 59.42. Result ID: 6993501d3a6c314849151c59.

[Reporting paper](<https://www.sota2.com/research/paper/68fa74541e209131cea611b9.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6da91e209131cea3b45b.md>)

Recorded evaluation configuration: {"Evaluation Protocol":"Zero-shot","Pre-training Dataset":"2x larger than standard"}. This record does not assert independent reproduction.

### MRM-VIT

Accuracy: 58.26. Result ID: 6993501d3a6c314849151c57.

[Reporting paper](<https://www.sota2.com/research/paper/68fa74541e209131cea611b9.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6e8b1e209131cea40752.md>)

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

### GLORIA-R50

Accuracy: 57.56. Result ID: 6993501d3a6c314849151c53.

[Reporting paper](<https://www.sota2.com/research/paper/68fa74541e209131cea611b9.md>)

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

### MedCLIP-Swin

Accuracy: 57.46. Result ID: 6993501d3a6c314849151c5b.

[Reporting paper](<https://www.sota2.com/research/paper/68fa74541e209131cea611b9.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6da91e209131cea3b45b.md>)

Recorded evaluation configuration: {"Evaluation Protocol":"Full-finetuning","Pre-training Dataset":"2x larger than standard"}. This record does not assert independent reproduction.

### MGCA-VIT+

Accuracy: 56.96. Result ID: 6993501d3a6c314849151c56.

[Reporting paper](<https://www.sota2.com/research/paper/68fa74541e209131cea611b9.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6d9e1e209131cea3ae0a.md>)

[Implementation](<https://github.com/fuying-wang/MGCA>)

Recorded evaluation configuration: {"Evaluation Protocol":"Full-finetuning","Pre-training Dataset":"2x larger than standard"}. This record does not assert independent reproduction.

### MRM-VIT

Accuracy: 56.56. Result ID: 6993501d3a6c314849151c58.

[Reporting paper](<https://www.sota2.com/research/paper/68fa74541e209131cea611b9.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6e8b1e209131cea40752.md>)

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

### ImageNet-ViT

Accuracy: 56.46. Result ID: 6993501d3a6c314849151c4b.

[Reporting paper](<https://www.sota2.com/research/paper/68fa74541e209131cea611b9.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa72181e209131cea546c3.md>)

[Implementation](<https://github.com/facebookresearch/dinov2>)

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

### MedCLIP-Swin

Accuracy: 54.55. Result ID: 6993501d3a6c314849151c5a.

[Reporting paper](<https://www.sota2.com/research/paper/68fa74541e209131cea611b9.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6da91e209131cea3b45b.md>)

Recorded evaluation configuration: {"Evaluation Protocol":"Linear-probing","Pre-training Dataset":"2x larger than standard"}. This record does not assert independent reproduction.

### GLORIA-R50

Accuracy: 51.65. Result ID: 6993501d3a6c314849151c52.

[Reporting paper](<https://www.sota2.com/research/paper/68fa74541e209131cea611b9.md>)

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

### MGCA-VIT+

Accuracy: 51.35. Result ID: 6993501d3a6c314849151c54.

[Reporting paper](<https://www.sota2.com/research/paper/68fa74541e209131cea611b9.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6d9e1e209131cea3ae0a.md>)

[Implementation](<https://github.com/fuying-wang/MGCA>)

Recorded evaluation configuration: {"Evaluation Protocol":"Zero-shot","Pre-training Dataset":"2x larger than standard"}. This record does not assert independent reproduction.

### ConVIRT-R50

Accuracy: 47.7. Result ID: 6993501d3a6c314849151c50.

[Reporting paper](<https://www.sota2.com/research/paper/68fa74541e209131cea611b9.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6d741e209131cea3a0a1.md>)

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

### CLIP-VIT

Accuracy: 44.84. Result ID: 6993501d3a6c314849151c4e.

[Reporting paper](<https://www.sota2.com/research/paper/68fa74541e209131cea611b9.md>)

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

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

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

### GLORIA-R50

Accuracy: 43.28. Result ID: 6993501d3a6c314849151c51.

[Reporting paper](<https://www.sota2.com/research/paper/68fa74541e209131cea611b9.md>)

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

### ConVIRT-R50

Accuracy: 41.88. Result ID: 6993501d3a6c314849151c4f.

[Reporting paper](<https://www.sota2.com/research/paper/68fa74541e209131cea611b9.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6d741e209131cea3a0a1.md>)

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

### ImageNet-ViT

Accuracy: 38.54. Result ID: 6993501d3a6c314849151c4a.

[Reporting paper](<https://www.sota2.com/research/paper/68fa74541e209131cea611b9.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa72181e209131cea546c3.md>)

[Implementation](<https://github.com/facebookresearch/dinov2>)

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

### CLIP-VIT

Accuracy: 35.34. Result ID: 6993501d3a6c314849151c4d.

[Reporting paper](<https://www.sota2.com/research/paper/68fa74541e209131cea611b9.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.

### MGCA-VIT+

Accuracy: 26.63. Result ID: 6993501d3a6c314849151c55.

[Reporting paper](<https://www.sota2.com/research/paper/68fa74541e209131cea611b9.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6d9e1e209131cea3ae0a.md>)

[Implementation](<https://github.com/fuying-wang/MGCA>)

Recorded evaluation configuration: {"Evaluation Protocol":"Linear-probing","Pre-training Dataset":"2x larger than standard"}. This record does not assert independent reproduction.

### Random-ViT

Accuracy: 20.62. Result ID: 6993501d3a6c314849151c48.

[Reporting paper](<https://www.sota2.com/research/paper/68fa74541e209131cea611b9.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa72181e209131cea546c3.md>)

[Implementation](<https://github.com/facebookresearch/dinov2>)

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

### Random-ViT

Accuracy: 20.32. Result ID: 6993501d3a6c314849151c49.

[Reporting paper](<https://www.sota2.com/research/paper/68fa74541e209131cea611b9.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa72181e209131cea546c3.md>)

[Implementation](<https://github.com/facebookresearch/dinov2>)

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

### CLIP-VIT

Accuracy: 20.16. Result ID: 6993501d3a6c314849151c4c.

[Reporting paper](<https://www.sota2.com/research/paper/68fa74541e209131cea611b9.md>)

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

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

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.
