1-of-100 Accuracy: 30.6%
Papers with Code · CC BY-SA 4.0
Conversational Response Selection
Conversational response selection refers to the task of identifying the most relevant response to a given input sentence from a collection of sentences.
논문
1-of-100 Accuracy: 71.8%
Accuracy: 36.93
MAP: 0.651MRR: 0.687P@1: 0.510R10@1: 0.328
Accuracy: 60.42
MAP: 0.648MRR: 0.688P@1: 0.518R10@1: 0.327
Accuracy: 40.83
MAP: 0.591MRR: 0.633P@1: 0.454R10@1: 0.280
1-of-100 Accuracy: 19.3%
MAP: 0.639MRR: 0.681P@1: 0.514R10@1: 0.330
MAP: 0.625MRR: 0.664P@1: 0.499R10@1: 0.318
MAP: 0.644MRR: 0.680P@1: 0.512R10@1: 0.324
MAP: 0.570MRR: 0.615P@1: 0.433R10@1: 0.262
MAP: 0.551MRR: 0.599P@1: 0.421R10@1: 0.243
Accuracy: 63.64
MAP: 0.587MRR: 0.632P@1: 0.470R10@1: 0.295
MAP: 0.550MRR: 0.601P@1: 0.427R10@1: 0.254
MAP: 0.573MRR: 0.621P@1: 0.444R10@1: 0.269
MAP: 0.608MRR: 0.650P@1: 0.475R10@1: 0.299
R@1: 31.0R@10: 78.8R@50: 97.8
MAP: 0.529MRR: 0.569P@1: 0.397R10@1: 0.233
1-of-100 Accuracy: 47.7%