TY  - RPRT
TI  - Multi-Agent Task-Oriented Dialog Policy Learning with Role-Aware Reward Decomposition
AU  - Ryuichi Takanobu
AU  - Runze Liang
AU  - Minlie Huang
PY  - 2020
UR  - https://arxiv.org/abs/2004.03809
ID  - 2004.03809
ER  -