Aspect-Level Sentiment-Controllable Review Generation with Mutual Learning Framework


  • Huimin Chen Tsinghua University
  • Yankai Lin Wechat Tencent
  • Fanchao Qi Tsinghua University
  • Jinyi Hu Tsinghua University
  • Peng Li Wechat Tencent
  • Jie Zhou Wechat Tencent
  • Maosong Sun Tsinghua University




Review generation, aiming to automatically generate review text according to the given information, is proposed to assist in the unappealing review writing. However, most of existing methods only consider the overall sentiments of reviews and cannot achieve aspect-level sentiment control. Even though some previous studies attempt to generate aspect-level sentiment-controllable reviews, they usually require large-scale human annotations which are unavailable in the real world. To address this issue, we propose a mutual learning framework to take advantage of unlabeled data to assist the aspect-level sentiment-controllable review generation. The framework consists of a generator and a classifier which utilize confidence mechanism and reconstruction reward to enhance each other. Experimental results show our model can achieve aspect-sentiment control accuracy up to 88% without losing generation quality.




How to Cite

Chen, H., Lin, Y., Qi, F., Hu, J., Li, P., Zhou, J., & Sun, M. (2021). Aspect-Level Sentiment-Controllable Review Generation with Mutual Learning Framework. Proceedings of the AAAI Conference on Artificial Intelligence, 35(14), 12639-12647. Retrieved from



AAAI Technical Track on Speech and Natural Language Processing I