Personal profile
Research profile
I am a PhD Fellow in the AI:SECURITY and NLP Labs at Aalborg University, supported by Aalborg University’s AI:X project. I am supervised by Prof. Johannes Bjerva and Asst. Prof. Qiongxiu (Jane) Li.
My research interests lie in neural network memorization, AI security and privacy, and multilingual LLMs.
Education/Academic qualification
Computer Science and Engineering, MSc., Polytechnic University of Milan
Sept 2020 → Apr 2024
Computer Science and Technology, BSc., Beijing University of Technology
Sept 2016 → Jul 2020
Keywords
- Computer Science
- LLMs
- Security and Privacy
- AI
- Multilingual NLP
Fingerprint
- 1 Similar Profiles
Projects
- 2 Active
-
AI:SECURITY
Bjerva, J. (PI), Li, Q. (PI), Bakka, J. M. (Project Participant), Luo, X. (Project Participant), Yu, W. (Project Participant) & Wisniewski, R. (Other)
01/08/2025 → 31/12/2029
Project: Research
-
Multilingual Modelling for Resource-Poor Languages
Bjerva, J. (PI), Lent, H. C. (Project Participant), Chen, Y. (Project Participant), Ploeger, E. (Project Participant), Fekete, M. R. (Project Participant), Lavrinovics, E. (Project Participant) & Luo, X. (Project Participant)
01/09/2022 → 31/08/2026
Project: Research
-
Shared Path: Unraveling Memorization in Multilingual LLMs through Language Similarities
Luo, X., Chen, Y., Bjerva, J. & Li, Q., Nov 2025, Proceedings of the 2025 Conference on Empirical Methods in Natural Language Processing: EMNLP 2025. Christodoulopoulos, C., Chakraborty, T., Rose, C. & Peng, V. (eds.). Association for Computational Linguistics (ACL), p. 19372–19388Research output: Contribution to book/anthology/report/conference proceeding › Article in proceeding › Research › peer-review
Open AccessFile8 Downloads (Pure) -
APEX: Probing Neural Networks via Activation Perturbation
Ren, T., Luo, X. & Li, Q., 3 Feb 2026, arXiv, 23 p.Research output: Working paper/Preprint › Preprint
Open Access -
Characterizing Memorization in Diffusion Language Models: Generalized Extraction and Sampling Effects
Luo, X., Yu, W., Li, Q. & Bjerva, J., 2 Mar 2026, arXiv, 21 p.Research output: Working paper/Preprint › Preprint
Open Access -
Do LLMs Really Memorize Personally Identifiable Information? Revisiting PII Leakage with a Cue-Controlled Memorization Framework
Luo, X., Chen, Y., Li, Q. & Bjerva, J., 7 Jan 2026, Accepted/In Press: Association for Computational Linguistics (ACL), 20 p.Research output: Working paper/Preprint › Working paper › Research › peer-review
Open AccessFile19 Downloads (Pure) -
Demem: Privacy-Enhanced Robust Adversarial Learning via De-Memorization
Luo, X. & Li, Q., 3 Sept 2025, 2025 IEEE 35th International Workshop on Machine Learning for Signal Processing (MLSP): Signal Processing in the Age of Lorge Language Models, MLSP 2025. IEEE (Institute of Electrical and Electronics Engineers), 6 p. 11204331. (IEEE Workshop on Machine Learning for Signal Processing).Research output: Contribution to book/anthology/report/conference proceeding › Article in proceeding › Research › peer-review