Open-domain dialogue systems aim to interact with humans through natural language texts in an open-ended fashion. However, the widely successful neural networks may not work well for dialogue systems, as they tend to generate generic responses. In this work, we propose an Equal-size Hard Expectation–Maximization (EqHard-EM) algorithm to train a multi-decoder model for diverse dialogue generation. Our algorithm assigns a sample to a decoder in a hard manner and additionally imposes an equal-assignment constraint to ensure that all decoders are well-trained. We provide detailed theoretical analysis to justify our approach. Further, experiments on two large-scale, open-domain dialogue datasets verify that our EqHard-EM algorithm generates high-quality diverse responses.

Bibtex

@misc{https://doi.org/10.48550/arxiv.2209.14627,
doi = {10.48550/ARXIV.2209.14627},
url = {https://arxiv.org/abs/2209.14627},
author = {Wen, Yuqiao and Hao, Yongchang and Cao, Yanshuai and Mou, Lili},
keywords = {Computation and Language (cs.CL), Artificial Intelligence (cs.AI), Machine Learning (cs.LG), FOS: Computer and information sciences, FOS: Computer and information sciences, I.2.7; I.2.6},
title = {An Equal-Size Hard EM Algorithm for Diverse Dialogue Generation},
publisher = {arXiv},
year = {2022},
copyright = {arXiv.org perpetual, non-exclusive license}
}

Related Research