Structure Understanding on TSR
79.1AccuracyDiVA-Former
Evaluation Results
| Method | Links | |
|---|---|---|
| DiVA-FormerModality=V+T2026.03 | 79.1 | |
| AdapterModality=V+T2026.03 | 74.3 | |
| AdapterModality=Vision2026.03 | 73.6 | |
| ResamplerModality=Vision2026.03 | 65.4 | |
| ResamplerModality=V+T2026.03 | 65.3 | |
| ResamplerModality=Text2026.03 | 65.1 | |
| AdapterModality=Text2026.03 | 47 | |
| DirectModality=V+T2026.03 | 15.1 | |
| DirectModality=Vision2026.03 | 14.8 | |
| DirectModality=Text2026.03 | 14.2 |