F1: 17.8BLEU-4: 1ROUGE-L: 19.3
Papers with Code · CC BY-SA 4.0
Visual Dialog
Visual Dialog requires an AI agent to hold a meaningful dialog with humans in natural, conversational language about visual content. Specifically, given an image, a dialog history, and a follow-up question about the image, the task is to answer the question.
논문
MRR: 0.7124Mean Rank: 2.96R@1: 58.28R@10: 94.45
MRR: 64.1Mean Rank: 4.45R@1: 50.92R@10: 88.81
MRR: 0.5965Mean Rank: 5.46R@1: 45.55R@10: 85.37
MRR: 63.98Mean Rank: 4.47R@1: 50.29R@10: 88.81
MRR: 62.22Mean Rank: 4.81R@1: 48.48R@10: 87.59
MRR: 66.38Mean Rank: 4.04R@1: 53.33R@10: 90.38
MRR: 62.94Mean Rank: 4.17R@1: 48.64R@10: 89.94
MRR: 68.92Mean Rank: 3.39R@1: 55.16R@10: 92.95
MRR: 57.88Mean Rank: 5.84R@1: 43.51R@10: 83.96
MRR: 0.6756Mean Rank: 3.75R@1: 54.64R@10: 91.48
MRR: 0.6792Mean Rank: 3.97R@1: 54.76R@10: 90.68
MRR: 0.6765Mean Rank: 3.73R@1: 54.65R@10: 91.47
MRR: 0.6285Mean Rank: 4.57R@1: 48.95R@10: 88.36
MRR: 0.6634Mean Rank: 3.93R@1: 52.71R@10: 90.73
MRR: 62.42Mean Rank: 4.70R@1: 48.55R@10: 87.75
Mean Rank: 4.86R@1: 48.53R@10: 87.43R@5: 78.66