Improve Interpretability of Neural Networks via Sparse Contrastive Coding.

Junhong Liu, Yijie Lin, Liang Jiang, Jia Liu, Zujie Wen, Xi Peng

Browse the full EMNLP paper archive.