TY - JOUR
AU - 
AB - 1,2,3,# 1,2,3,# 1,2,3, 4 4 Zhuosheng Zhang , Siru Ouyang , Hai Zhao , Masao Utiyama , Eiichiro Sumita Department of Computer Science and Engineering, Shanghai Jiao Tong University Key Laboratory of Shanghai Education Commission for Intelligent Interaction and Cognitive Engineering, Shanghai Jiao Tong University MoE Key Lab of Artificial Intelligence, AI Institute, Shanghai Jiao Tong University National Institute of Information and Communications Technology (NICT), Kyoto, Japan {zhangzs,oysr0926}@sjtu.edu.cn,zhaohai@cs.sjtu.edu.cn {mutiyama,eiichiro.sumita}@nict.go.jp Abstract with people according to the dialogue states, and completes specific tasks, such as ordering meals Conversational machine reading (CMR) re- (Liu et al., 2013) and air tickets (Price, 1990). In quires machines to communicate with humans real-world scenario, annotating data such as in- through multi-turn interactions between two tents and slots is expensive. Inspired by the studies salient dialogue states of decision making and of reading comprehension (Rajpurkar et al., 2016, question generation processes. In open CMR 2018; Zhang et al., 2020c, 2021), there appears a settings, as the more realistic scenario, the re- trieved background knowledge would be noisy, more general task — conversational machine read- which results in severe challenges in the in- ing (CMR) (Saeidi et al., 2018): given the inquiry, formation transmission. Existing studies com- 
TI - Smoothing Dialogue States for Open Conversational Machine Reading
JF - Proceedings of the 2021 Conference on Empirical Methods in Natural Language Processing
DO - 10.18653/v1/2021.emnlp-main.299
DA - 2021-01-01
UR - https://www.deepdyve.com/lp/unpaywall/smoothing-dialogue-states-for-open-conversational-machine-reading-0B6woBQNdv
DP - DeepDyve
ER -