| A gradient control method for backdoor attacks on parameter-efficient tuning N Gu, P Fu, X Liu, Z Liu, Z Lin, W Wang Proceedings of the 61st Annual Meeting of the Association for Computational …, 2023 | 7 | 2023 |
| DE-CO: A Two-Step Spelling Correction Model for Combating Adversarial Typos Z Liu, F Wang, Z Lin, L Wang, Z Yin 2020 IEEE Intl Conf on Parallel & Distributed Processing with Applications …, 2020 | 6 | 2020 |
| Cost-eff: Collaborative optimization of spatial and temporal efficiency with slenderized multi-exit language models B Shen, Z Lin, Y Liu, Z Liu, L Wang, W Wang arXiv preprint arXiv:2210.15523, 2022 | 3 | 2022 |
| Maximum entropy loss, the silver bullet targeting backdoor attacks in pre-trained language models Z Liu, B Shen, Z Lin, F Wang, W Wang Findings of the Association for Computational Linguistics: ACL 2023, 3850-3868, 2023 | 2 | 2023 |
| MACROBERT: Maximizing Certified Region of BERT to Adversarial Word Substitutions F Wang, Z Lin, Z Liu, M Zheng, L Wang, D Zha International Conference on Database Systems for Advanced Applications, 253-261, 2021 | 2 | 2021 |
