REM-Net: Recursive Erasure Memory Network for Commonsense Evidence Refinement


  • Yinya Huang Sun Yat-Sen University
  • Meng Fang Tencent
  • Xunlin Zhan Sun Yat-Sen University
  • Qingxing Cao Sun Yat-Sen University
  • Xiaodan Liang Sun Yat-sen University


Common-Sense Reasoning, Question Answering, Applications


When answering a question, people often draw upon their rich world knowledge in addition to the particular context. While recent works retrieve supporting facts/evidence from commonsense knowledge bases to supply additional information to each question, there is still ample opportunity to advance it on the quality of the evidence. It is crucial since the quality of the evidence is the key to answering common- sense questions, and even determines the upper bound on the QA systems’ performance. In this paper, we propose a recursive erasure memory network (REM-Net) to cope with the quality improvement of evidence. To address this, REM-Net is equipped with a module to refine the evidence by recursively erasing the low-quality evidence that does not explain the question answering. Besides, instead of retrieving evidence from existing knowledge bases, REM-Net leverages a pre-trained generative model to generate candidate evidence customized for the question. We conduct experiments on two commonsense question answering datasets, WIQA and CosmosQA. The results demonstrate the performance of REM- Net and show that the refined evidence is explainable.




How to Cite

Huang, Y., Fang, M., Zhan, X., Cao, Q., & Liang, X. (2021). REM-Net: Recursive Erasure Memory Network for Commonsense Evidence Refinement. Proceedings of the AAAI Conference on Artificial Intelligence, 35(7), 6375-6383. Retrieved from



AAAI Technical Track on Knowledge Representation and Reasoning