no code implementations • 30 Apr 2020 • Jiayi Zhang, Chongyang Tao, Zhenjing Xu, Qiaojing Xie, Wei Chen, Rui Yan
Aiming at generating responses that approximate the ground-truth and receive high ranking scores from the discriminator, the two generators learn to generate improved highly relevant responses and competitive unobserved candidates respectively, while the discriminative ranker is trained to identify true responses from adversarial ones, thus featuring the merits of both generator counterparts.