Visual Question Answering on CulturalVQA
0.354ECEMMBoundary
Evaluation results
| Method | Links | |||||
|---|---|---|---|---|---|---|
| MMBoundaryBase Model=Qwen2VL 7B2025.05 | 0.354 | 0.385 | 43.7 | 63.4 | — | |
| SaySelfBase Model=Qwen2VL 7B2025.05 | 0.397 | 0.426 | 35.2 | 58.6 | — | |
| RCEBase Model=Qwen2VL 7B2025.05 | 0.436 | 0.475 | 42.5 | 59.7 | — | |
| MultisampleBase Model=Qwen2VL 7B2025.05 | 0.443 | 0.512 | 33.5 | 51.3 | — | |
| DRLBase Model=Qwen2VL 7B2025.05 | 0.465 | 0.493 | 38.4 | 55.2 | — | |
| Conf-CSRBase Model=Qwen2VL 7B2025.05 | 0.482 | 0.503 | 39.4 | 56.2 | — | |
| Conf-CSR2025.05 | — | — | — | — | 0.4931 | |
| DRL2025.05 | — | — | — | — | 0.4576 | |
| MMBoundary2025.05 | — | — | — | — | 0.7108 | |
| Multisample2025.05 | — | — | — | — | 0.4904 | |
| RCE2025.05 | — | — | — | — | 0.6059 | |
| SaySelf2025.05 | — | — | — | — | 0.6261 |