About me
Bio
I am currently a PhD student at City University of Hong Kong (CityU), advised by Prof. Xiangyu Zhao. I completed my MPhil at University of Science and Technology of China (USTC), advised by Prof. Qi Liu. Before that, I graduated from Hefei University of Technology (HFUT), advised by Prof. Le Wu.
Research areas that I’m interested in (but not limited to) are Trustworthy LLM, Post-training of LLM and GenRec.
I am looking for LLM-related internships worldwide starting in about six months. Feel free to contact me if you have opportunities or would like to connect.
Education
PhD in Data Science
Department of Data Science, City University of Hong Kong (CityU HK)
Sep. 2024 – Now
MPhil in Computer Technology
School of Computer Science, University of Science and Technology of China (USTC)
Sep. 2022 – July. 2025
Bachelor’s in Computer Science
School of Computer Science and Information Engineering, Hefei University of Technology (HFUT)
Sep. 2018 – July. 2022
Publications
Preprint
AdaSwitch: Balancing Exploration and Guidance in Knowledge Distillation via Adaptive Switching
Jingyu Peng, Maolin Wang, Hengyi Cai, Yuchen Li, Kai Zhang, Shuaiqiang Wang, Dawei Yin, Xiangyu Zhao
[PDF]
MOSAIC: Composable Safety Alignment with Modular Control Tokens
Jingyu Peng, Hongyu Chen, Jiancheng Dong, Maolin Wang, Wenxi Li, Yuchen Li, Kai Zhang, Xiangyu Zhao
[PDF]
Conferences
Logic Jailbreak: Efficiently Unlocking LLM Safety Restrictions Through Formal Logical Expression
Jingyu Peng, Maolin Wang, Nan Wang, Jiatong Li, Yuchen Li, Yuyang Ye, Wanyu Wang, Pengyue Jia, Kai Zhang, Xiangyu Zhao
The 64rd Annual Meeting of the Association for Computational Linguistics (ACL 2026, CCF-A)
[PDF]
Stepwise Reasoning Error Disruption Attack of LLMs
Jingyu Peng, Maolin Wang, Xiangyu Zhao, Kai Zhang, Wanyu Wang, Pengyue Jia, Qidong Liu, Ruocheng Guo, Qi Liu
The 63rd Annual Meeting of the Association for Computational Linguistics (ACL 2025, CCF-A)
[PDF]
Towards Few-shot Self-explaining Graph Neural Networks
Jingyu Peng, Qi Liu, Linan Yue, Zaixi Zhang, Kai Zhang and Yunhao Sha
Joint European Conference on Machine Learning and Knowledge Discovery in Databases 2024 (CCF-B)
[PDF]
Meta-learned ID Embeddings for Online Inductive Recommendation
Jingyu Peng, Le Wu, Peijie Sun & Meng Wang
China Conference on Information Retrieval 2021
[PDF]
On the Continued Value of Universal Dependencies in the Era of Large Language Models
Wenxi Li, Jingyu Peng
The 64rd Annual Meeting of the Association for Computational Linguistics (ACL 2026, CCF-A)
[PDF]
Beyond Homophily: Robust Graph Anomaly Detection via Neural Sparsification
Zheng Gong, Guifeng Wang, Ying Sun, Qi Liu, Yuting Ning, Hui Xiong, Jingyu Peng
The 32nd International Joint Conference on Artificial Intelligence (IJCAI-23, CCF-A)
[PDF]
Internship
Baidu Search
- Research on LLM industry-specific AI search, with a particular emphasis on Knowledge Distillation.
Alibaba-Inc
- Research on multimodal search algorithms for Goofish(闲鱼) platform.
- Research on holistic modeling methods for product UI in search interfaces. June. 2024 - Sep. 2024
Alipay
- Participate in the development of Ray core.
- Participate in the development of Mars.
- Study the operating efficiency of various machine learning algorithms(XGBoost, Lightgbm) on Mars & Ray June. 2021 - Aug. 2021
IFLYTEK
- Responsible for the research of Job Competency Perception Diagnosis.
- Participate in the construction of dataset. May. 2022 - Jul. 2022
Service
- Reviewer: WWW 2025, ICANN 2025, CIKM 2025, ARR May,Oct 2025, AAAI 2026
- Journal Reviewer: Applied Intelligence
Competitions
U.S. Patent Phrase to Phrase Matching
- It’s my first time to participate in Kaggle competition. I got a silver medal in U.S. Patent Phrase to Phrase Matching with my teammate Zheng Gong. The solution is available. May. 2022
Optiver Trading at the Close
- Silver medal (top 2%)
Awards
Awarded as Outstanding Graduate in HFUT. June. 2022
Awarded First Class Scholarship in USTC. Sep. 2024, Sep. 2022
Miscellaneous
I love extreme sports including Motorcycle, Car Racing and Snowboarding.