I am a fourth-year Ph.D. student advised by Prof. Kenji Kawaguchi at the School of Computing (SoC), National University of Singapore (NUS). I am also currently an intern with Prof. Dawn Song’s team at UC Berkeley. Previously, I graduated from Tsinghua University with a B.S. degree in Electronic Engineering. I’ve had the fortune to work with Prof. Dongmei Li at Tsinghua University. Afterwards, I joined Center for Speech and Language Technologies (CSLT) as a research intern with Dr. Lantian Li and Prof. Dong Wang. Then I became an intern in ASR Oteam, Tencent Inc. in Beijing and did research in ASR and Multimodal Learning, organizing ICPR MSR 2022 with Dr. Jian Kang, etc.
My research focuses on large language models, with an emphasis on the security and reliability of LLM-based agents and the foundations of reasoning in modern language models. I study how agentic systems fail under adversarial or ambiguous conditions, and investigate the internal mechanisms that enable or hinder robust multi-step reasoning, with the goal of improving safety, trustworthiness, and generalization.
🔥 News
- [Feb 2026] We open-sourced a high-quality strategy extraction and reasoning-guidance pipeline, enabling structured reasoning supervision and execution: strategy-execute-pipeline
- [Feb 2026] We released HM-ReasoningBench, the first competitive dataset pairing model-generated solutions with human reasoning traces: HM-ReasoningBench
- [Jan 2026] Our paper, CARE: Contextual Adaptation of Recommenders for LLM-based Conversational Recommendation, was accepted by ECIR 2026!
📝 Publications
[1] CARE: Contextual Adaptation of Recommenders for LLM-based Conversational Recommendation Chuang Li, Weida Liang, Yang Deng, Hengchang Hu, See-Kiong Ng, Min-Yen Kan, Haizhou Li ECIR 2026 [PDF]
[2] From Harm to Help: Turning Reasoning In-Context Demos into Assets for Reasoning LMs Weida Liang, Haonan Wang, Zihang Fu, Nie Zheng, Yifan Zhang, Yao Tong, Tongyao Zhu, Hao Jiang, Chuang Li, Jiaying Wu, Kenji Kawaguchi arXiv 2025 [PDF]
[3] PromptArmor: Simple yet Effective Prompt Injection Defenses Tianneng Shi, Kaijie Zhu, Zhun Wang, Yuqi Jia, Will Cai, Weida Liang, Haonan Wang, Hend Alzahrani, etc. arXiv 2025 [PDF]
[4] ICPR 2022 Challenge on Multi-Modal Subtitle Recognition Shan Huang, Shen Huang, Li Lu, Pengfei Hu, Lijuan Wang, Xiang Wang*, Jian Kang, Weida Liang, etc. ICPR 2022 [Project Page]
[5] Enhanced Exemplar Autoencoder with Cycle Consistency Loss in Any-to-One Voice Conversion Weida Liang, Lantian Li, Dong Wang, Wenqiang Du arXiv 2022 [PDF] [Code] [Project Page]
📖 Education
2022.8 - Present
Ph.D. Student in Computer Science
Advisor: Prof. Kenji Kawaguchi
💼 Research Experience
2021.8 - 2022.4
Advisor: Prof. Dong Wang
💻 Work Experience
2025.7 - Present
TikTok Live
2022.3 - 2022.6
ASR Oteam
✉️ Contact
- Email: weida_liang[at]u.nus.edu
- GitHub: lwd17
- Google Scholar: Weida Liang
- WeChat: Click to view
