Text Retrieval On Image Chat
평가 지표
R@1
R@5
Sum(R@1,5)
평가 결과
이 벤치마크에서 각 모델의 성능 결과
모델 이름 | R@1 | R@5 | Sum(R@1,5) | Paper Title | Repository |
|---|---|---|---|---|---|
| TransResNet | 50.3 | 75.4 | 125.7 | Image Chat: Engaging Grounded Conversations | |
| VLMo | 46.8 | 67.5 | 114.3 | VLMo: Unified Vision-Language Pre-Training with Mixture-of-Modality-Experts | |
| PaCE | 51.9 | 76.8 | 128.7 | PaCE: Unified Multi-modal Dialogue Pre-training with Progressive and Compositional Experts |
0 of 3 row(s) selected.