MS-CXR
Benchmarks
| Task Name | Dataset Name | SOTA Result | Trend | |
|---|---|---|---|---|
| Interval-change classification | MS-CXR-T | Average Macro Accuracy68.5 | 24 | |
| Multi-label diagnostic classification | MS-CXR-T | Consolidation Score77.7 | 21 | |
| Phrase Grounding | MS-CXR | Atelectasis Accuracy0.8449 | 15 | |
| Temporal Image Classification | MS-CXR-T (test) | Macro Acc (Pleural Effusion)68.1 | 14 | |
| Feature Attribution | MS-CXR text (test) | Conf. Drop (%)1.16 | 13 | |
| Visual Grounding | MS-CXR (test) | mAP@0.587.9 | 8 | |
| Attribution | MS-CXR | Confidence Drop0.51 | 7 | |
| Grounding | MS-CXR (test) | Pointing Game87.2 | 6 | |
| Feature Attribution | MS-CXR image (test) | Confidence Drop (%)0.0055 | 6 | |
| Referring Expression Grounding | MS-CXR (test) | Accuracy69.9 | 3 | |
| Referring Expression Grounding | MS-CXR MAIRA-2 7B (test) | mIoU57.8 | 1 | |
| Phrase Grounding | MS-CXR (test) | Atelectasis Accuracy- | 0 | |
| Phrase grounding | MS-CXR 1.0 | Metric- | 0 |
Showing 13 of 13 rows