Phrase-Level Action Reinforcement Learning for Neural Dialog Response Generation.

Takato Yamazaki, Akiko Aizawa

Browse the full ACL paper archive.