Rethinking Stealthiness of Backdoor Attack against NLP Models

Published in ACL, 2021

**Wenkai Yang**, Yankai Lin, Peng Li, Jie Zhou, Xu Sun [[code](https://github.com/lancopku/SOS)]