🎈 About Me

I am currently a third-year Ph.D. student at AI Thrust, Information Hub, The Hong Kong University of Science and Technology (Guangzhou) supervised by Prof. Li Liu and Prof. Hui Xiong. Previously, I received my master degree from City University of Hong Kong, advised by Prof. Xiangyu Zhao.

Research Interests

• AI Security / Safety
• Red-teaming Agent
• Backdoor Learning
• Recommendation

📝 Selected Papers

(* indicates equal contribution, # indicates corresponding author)

Publications

NeurIPS 2025 D&B

BackdoorDM: A Comprehensive Benchmark for Backdoor Learning on Diffusion Model

Weilin Lin^*, Nanjun Zhou^*, Yanyun Wang, Jianze Li, Hui Xiong, Li Liu^#

• The first comprehensive benchmark for backdoor learning on diffusion models.
• Propose a unified attack formulation and a systematic target taxonomy.
• Support 9 diffusion backdoor attacks, 5 defense methods, and 3 visualization tools.

The Thirty-Ninth Annual Conference on Neural Information Processing Systems Datasets & Benchmarks Track (NeurIPS D&B), San Diego, California, USA, 2025

AAAI 2025 Oral

Fusing Pruned and Backdoored Models: Optimal Transport-based Data-free Backdoor Mitigation Oral Presentation (4.6%)

Weilin Lin, Li Liu^#, Jianze Li, Hui Xiong

• One of the few data-free defense strategies against backdoor attacks.
• First adaptation of OT and model fusion on backdoor defense.

The Thirty-Ninth AAAI Conference on Artificial Intelligence (AAAI-25), Philadelphia, Pennsylvania, USA, 2025

NeurIPS 2024

Unveiling and Mitigating Backdoor Vulnerabilities based on Unlearning Weight Changes and Backdoor Activeness

Weilin Lin, Li Liu^#, Shaokui Wei, Jianze Li, Hui Xiong

• New insights on unlearning weight change and backdoor activeness.
• Propose an effective defense strategy using reinitialization and fine-tuning.

Annual Conference on Neural Information Processing Systems (NeurIPS), Vancouver, Canada, 2024

• ICASSP 2025 Gradient Norm-based Fine-Tuning for Backdoor Defense in Automatic Speech Recognition, Nanjun Zhou^*, Weilin Lin^*, Li Liu^#.
• WWW 2023 Autodenoise: Automatic data instance denoising for recommendations, Weilin Lin, Xiangyu Zhao^#, Yejing Wang, Yuanshao Zhu, Wanyu Wang.
• KDD 2022 AdaFS: Adaptive feature selection in deep recommender system, Weilin Lin, Xiangyu Zhao^#, Yejing Wang, Tong Xu, Xian Wu.

Preprints

Arxiv 2025

SARSteer: Safeguarding Large Audio Language Models via Safe-Ablated Refusal Steering

Weilin Lin, Jianze Li, Hui Xiong, Li Liu^#

• The first inference-time defense framework specifically designed for Large Audio Language Models (LALMs).
• Proposes text-derived refusal steering to enforce refusal without manipulating audio inputs.
• Introduces decomposed safety space ablation to mitigate over-refusal on benign speech queries.

arXiv preprint, 2025

• Arxiv 2024 Segment anything for videos: A systematic survey, Chunhui Zhang, Yawen Cui, Weilin Lin, Guanjie Huang, Yan Rong, Li Liu^#, Shiguang Shan.
• Arxiv 2023 A comprehensive survey on segment anything model for vision and beyond, Chunhui Zhang, Li Liu^#, Yawen Cui, Guanjie Huang, Weilin Lin, Yiqian Yang, Yuehong Hu.

📖 Educations

• Ph.D., Artificial Intelligence, The Hong Kong University of Science and Technology (Guangzhou), September 2023 - Present
  • Advisors: Prof. Li Liu and Prof. Hui Xiong
• M.S.c, Multimedia Information Technology, City University of Hong Kong, August 2021 - October 2022
  • Advisor: Prof. Xiangyu Zhao
• B.S., Electronic Information Science and Technology, South China Normal University, September 2017 - July 2021

🧭 Experience

• 2025.12 - Present, Research Intern, WeChat Security Team, Tencent. Research on Red Teaming Agents.
• 2023.03 - 2023.09, Research Assistant, The Hong Kong University of Science and Technology (Guangzhou). Advisor: Prof. Li Liu.
• 2022.07 - 2023.03, Research Assistant, USAIL Group, HKUST Fok Ying Tung Research Institute / The Hong Kong University of Science and Technology (Guangzhou). Advisor: Prof. Hao Liu.
• 2021.09 - 2022.10, Research Assistant, Applied Machine Learning Lab (AML Lab), City University of Hong Kong. Advisor: Prof. Xiangyu Zhao.

🎖 Honors and Awards

• Outstanding Paper Award (Second Prize), The 15th Guangdong-Hong Kong-Macao Conference on Image and Graphics, December 2025. Awarded for the paper “BackdoorDM: A Comprehensive Benchmark for Backdoor Learning on Diffusion Model”.
• Best Student Paper Award Finalist in ICSR 2024.
• Full Postgraduate Scholarship, HKUST(GZ).

🍀 Services

Reviewer/External Reviewer

• The International Conference on Learning Representations (ICLR), 2026
• The International Conference on Machine Learning (ICML), 2025, 2026
• Annual Conference on Neural Information Processing Systems (NeurIPS), 2024, 2025
• Annual AAAI Conference on Artificial Intelligence (AAAI), 2024, 2025, 2026
• Conference on Research and Development in Information Retrieval (SIGIR), 2024