PLATO: Pre-trained Dialogue Generation Model With Discrete Latent Variable

Bao Siqi, He Huang, Wang Fan, Wu Hua, Wang Haifeng. Arxiv 2019

[Paper]    
Attention Mechanism Model Architecture RAG Reinforcement Learning Tools Training Techniques Transformer

Pre-training models have been proved effective for a wide range of natural language processing tasks. Inspired by this, we propose a novel dialogue generation pre-training framework to support various kinds of conversations, including chit-chat, knowledge grounded dialogues, and conversational question answering. In this framework, we adopt flexible attention mechanisms to fully leverage the bi-directional context and the uni-directional characteristic of language generation. We also introduce discrete latent variables to tackle the inherent one-to-many mapping problem in response generation. Two reciprocal tasks of response generation and latent act recognition are designed and carried out simultaneously within a shared network. Comprehensive experiments on three publicly available datasets verify the effectiveness and superiority of the proposed framework.

Similar Work