About me

I’m a Research Fellow at University College London (NLP group). I received my Ph.D. from Monash University (Australia), supervised by Prof. Reza Haffari and Dr. Mohammad Norouzi. My recent research lies in an intersection between deep learning and natural language processing, with an emphasis on robustness and security in NLP models.

Now, I’m highlighting:

• Security and safety in NLP models
• Robustness in NLP models

Recent News

• [September 2024 - Paper]: Our paper “Generative Models are Self-Watermarked: Declaring Model Authentication through Re-Generation” is accpeted by TMLR 2024.
• [August 2024 - Paper]: I will be presenting a tutorial titled “A Copyright War: Authentication for Large Language Models” at IJCAI 2024.
• [May 2024 - Paper]: 2 papers accepted at ACL (1x)/ACL Findings (1x) 2024
• [April 2024 - Paper]: Our paper “SEEP: Training Dynamics Grounds Latent Representation Search for Mitigating Backdoor Poisoning Attacks” is accpeted by TACL 2024.
• [March 2024 - Paper]: 2 papers accepted at NAACL 2024

Recent Research Highlights

• SEEP: Training Dynamics Grounds Latent Representation Search for Mitigating Backdoor Poisoning Attacks
  Xuanli He, Qiongkai Xu, Jun Wang, Benjamin Rubinstein, Trevor Cohn. In Transactions of the Association for Computational Linguistics. 2024.
• Here’s a Free Lunch: Sanitizing Backdoored Models with Model Merge
  Ansh Arora, Xuanli He, Maximilian Mozes, Srinibas Swain, Mark Dras, Qiongkai Xu. In Findings of the Association for Computational Linguistics ACL 2024. 2024.
• Backdoor Attacks on Multilingual Machine Translation
  Jun Wang, Qiongkai Xu, Xuanli He, Benjamin Rubinstein, Trevor Cohn. In Proceedings of the 2024 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies (Volume 1: Long Papers). 2024.
• Using Natural Language Explanations to Improve Robustness of In-context Learning
  Xuanli He, Yuxiang Wu, Oana-Maria Camburu, Pasquale Minervini, Pontus Stenetorp. In Proceedings of the 62nd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers). 2024.
• Mitigating Backdoor Poisoning Attacks through the Lens of Spurious Correlation
  Xuanli He, Qiongkai Xu, Jun Wang, Benjamin Rubinstein, Trevor Cohn. In Proceedings of the 2023 Conference on Empirical Methods in Natural Language Processing. 2023.
• Foiling Training-Time Attacks on Neural Machine Translation Systems
  Jun Wang, Xuanli He, Benjamin Rubinstein, Trevor Cohn. In Findings of the Association for Computational Linguistics: EMNLP 2022. 2022.
• Extracted BERT Model Leaks More Information than You Think!
  Xuanli He, Chen Chen, Lingjuan Lyu, Qiongkai Xu. In Proceedings of the 2022 Conference on Empirical Methods in Natural Language Processing. 2022.
• Protecting Intellectual Property of Language Generation APIs with Lexical Watermark
  Xuanli He, Qiongkai Xu, Lingjuan Lyu, Fangzhao Wu, Chenguang Wang. In Proceedings of the AAAI Conference on Artificial Intelligence. 2022.
• CATER: Intellectual Property Protection on Text Generation APIs via Conditional Watermarks
  Xuanli He, Qiongkai Xu, Yi Zeng, Lingjuan Lyu, Fangzhao Wu, Jiwei Li, Ruoxi Jia. In Proceedings of Advances in Neural Information Processing Systems. 2022.
• Generate, Annotate, and Learn: NLP with Synthetic Text
  Xuanli He, Islam Nassar, Jamie Kiros, Gholamreza Haffari, Mohammad Norouzi. In Transactions of the Association for Computational Linguistics. 2022.
• Model Extraction and Adversarial Transferability, Your BERT is Vulnerable!
  Xuanli He, Lingjuan Lyu, Lichao Sun, and Qiongkai Xu. In Proceedings of the 2021 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies. 2021.