1 code implementation • 15 Apr 2023 • Sibo Wei, Wenpeng Lu, Xueping Peng, Shoujin Wang, Yi-Fei Wang, Weiyu Zhang
Although existing works have attempted to utilize Seq2Seq, reinforcement learning, or contrastive learning to solve the problem, two challenges remain: how to correctly capture question focus to model its semantic intention, and how to obtain reliable datasets to fairly evaluate performance.