# PLATO-2: Towards Building an Open-Domain Chatbot via Curriculum Learning

**Authors:** Siqi Bao, Huang He, Fan Wang, Hua Wu, Haifeng Wang, Wenquan Wu, Zhen Guo, Zhibin Liu, Xinchao Xu  
**arXiv:** [2006.16779](https://arxiv.org/abs/2006.16779)  
**Date:** 2020-06-30  
**Summary page:** https://www.summarizepaper.com/fr/paper/2006.16779v1/

## Abstract

To build a high-quality open-domain chatbot, we introduce the effective training process of PLATO-2 via curriculum learning. There are two stages involved in the learning process. In the first stage, a coarse-grained generation model is trained to learn response generation under the simplified framework of one-to-one mapping. In the second stage, a fine-grained generation model and an evaluation model are further trained to learn diverse response generation and response coherence estimation, respectively. PLATO-2 was trained on both Chinese and English data, whose effectiveness and superiority are verified through comprehensive evaluations, achieving new state-of-the-art results.