Generalized Referring Expression Comprehension on gRefCOCO (val)
61.9Pr@F1ReLA
Evaluation Results
| Method | Links | |||
|---|---|---|---|---|
| ReLAVisual Encoder=Swin-B, Textual Encoder=BERT2026.01 | 61.9 | 56.37 | 96.32 | |
| ReLAVisual Encoder=ResNet-50, Textual Encoder=BERT2026.01 | 59.36 | 55.83 | 96.36 | |
| UNINEXTVisual Encoder=ResNet-50, Textual Encoder=BERT2026.01 | 58.19 | 50.58 | 96.52 | |
| MDETRVisual Encoder=ResNet-101, Textual Encoder=RoBERTa2026.01 | 42.69 | 36.27 | 99.4 | |
| VLTVisual Encoder=DarkNet-53, Textual Encoder=GRU2026.01 | 36.62 | 35.2 | 99.44 | |
| TransVGVisual Encoder=ResNet-101, Textual Encoder=BERT2026.01 | 30.96 | 31.18 | 99.5 | |
| MCNVisual Encoder=DarkNet-53, Textual Encoder=GRU2026.01 | 28.02 | 30.64 | 99.62 |