Jiayu Liu  (刘嘉聿)


Research Assistant Professor

Hong Kong Institute of AI for Science
City University of Hong Kong (CityU)

Email: jiayuliu@cityu.edu.hk
Office: City University of Hong Kong, Kowloon, Hong Kong SAR.

Biography

I am a Research Assistant Professor at the Hong Kong Institute of AI for Science, City University of Hong Kong (CityU). I received my Ph.D. from the School of Artificial Intelligence and Data Science, University of Science and Technology of China (USTC), where I was a member of the BASE Group of the State Key Laboratory of Cognitive Intelligence, co-supervised by Prof. Enhong Chen and Prof. Zhenya Huang. From January 2025 to January 2026, I was fortunate to visit Prof. Yee Whye Teh's group at the University of Oxford. I received my Bachelor degree in Applied Mathematics from USTC in July, 2020.

My research goal is to endow machines with human-like abilities — such as knowledge learning and reasoning, latent reasoning, and self-evolving — to resolve real-world problems. I am also interested in the application of AI to specific domains, especially AI4Math: how to make AI collaborate better with mathematicians and assist in mathematical theorem proving.

Experience & Education

Selected Publications & Preprints

[Google Scholar]

( *, Corresponding author;  †, Equal contribution )

Deep Thinking by Markov Chain of Continuous Thoughts
Jiayu Liu, Zhenya Huang, Xuan Yang, Tianyun Ji, Anya Sims, Hao Xu, Enhong Chen, Yee Whye Teh, Ning Miao.
arXiv preprint, 2026. [PDF] [Code]
UniCog: Uncovering Cognitive Abilities of LLMs through Latent Mind Space Analysis
Jiayu Liu†, Yinhe Long†, Zhenya Huang, Enhong Chen.
arXiv preprint, 2026. [PDF] [Code]
Verifying Large Language Models' Reasoning Paths via Correlation Matrix Rank
Jiayu Liu, Wei Dai, Zhenya Huang, Ning Miao, Enhong Chen.
arXiv preprint, 2026. [PDF] [Code]
Step-Level Sparse Autoencoder for Reasoning Process Interpretation
Xuan Yang†, Jiayu Liu†, Yuhang Lai, Hao Xu, Zhenya Huang, Ning Miao.
ICML'2026, PMLR 306, Seoul, South Korea, 2026. [PDF] [Code]
Foundation of Intelligence: Review of Math Word Problems from Human Cognition Perspective
Zhenya Huang, Jiayu Liu, Xin Lin, Zhiyuan Ma, Shangzi Xue, Tong Xiao, Qi Liu, Yee Whye Teh, Enhong Chen.
arXiv preprint, 2025. [PDF]
Enhancing Mathematical Reasoning Through Autonomously Learning Knowledge
Jiayu Liu, Zhenya Huang, Enhong Chen, Qi Liu, Hongke Zhao, Xin Lin, Jing Sha, Shijin Wang.
IEEE Transactions on Pattern Analysis and Machine Intelligence (TPAMI), 48(3), 3100-3118, 2026. [PDF] [Code]
Knowledge-Centered Dual-Process Reasoning for Math Word Problems with Large Language Models
Jiayu Liu, Zhenya Huang, Qi Liu, Zhiyuan Ma, Chengxiang Zhai, Enhong Chen.
IEEE Transactions on Knowledge and Data Engineering (TKDE), 37(6), 3457-3471, 2025. [PDF] [Code]
What Makes In-context Learning Effective for Mathematical Reasoning
Jiayu Liu, Zhenya Huang, Chaokun Wang, Xunpeng Huang, ChengXiang Zhai, Enhong Chen.
ICML'2025, PMLR 267:38708-38721, Vancouver, Canada, July 13-19, 2025. [PDF] [Code]
CogMath: Assessing LLMs' Authentic Mathematical Ability from a Human Cognitive Perspective
Jiayu Liu, Zhenya Huang, Wei Dai, Cheng Cheng, Jinze Wu, Jing Sha, Song Li, Qi Liu, Shijin Wang, Enhong Chen.
ICML'2025, PMLR 267:38692-38707, Vancouver, Canada, July 13-19, 2025. [PDF] [Code]
SocraticLM: Exploring Socratic Personalized Teaching with Large Language Models
Jiayu Liu, Zhenya Huang, Tong Xiao, Jing Sha, Jinze Wu, Qi Liu, Shijin Wang, Enhong Chen.
NeurIPS'2024 (Spotlight), 85693-85721, Vancouver, Canada, December 10-15, 2024. [PDF] [Code]
Guiding Mathematical Reasoning via Mastering Commonsense Formula Knowledge
Jiayu Liu, Zhenya Huang*, Zhiyuan Ma, Qi Liu, Enhong Chen*, Tianhuang Su, Haifeng Liu.
KDD'2023, 1477–1488, Long Beach, CA, USA, August 6-10, 2023. [PDF] [Code]
Learning by Applying: A General Framework for Mathematical Reasoning via Enhancing Explicit Knowledge Learning
Jiayu Liu, Zhenya Huang*, Chengxiang Zhai, Qi Liu.
AAAI'2023, 4497-4506, Washington, DC, USA, Feb.7-14, 2023. [PDF] [Code]
A Cognitive Solver with Autonomously Knowledge Learning for Reasoning Mathematical Answers
Jiayu Liu, Zhenya Huang*, Xin Lin, Qi Liu, Jianhui Ma, Enhong Chen.
ICDM'2022, 269-278, Orlando, FL, USA, November 28-December 1, 2022. [PDF] [Code]
A Probabilistic Framework for Temporal Cognitive Diagnosis in Online Learning Systems
Jiayu Liu, Fei Wang, Haiping Ma, Zhenya Huang, Qi Liu, Enhong Chen, Yu Su.
Journal of Computer Science and Technology (JCST), 38(6): 1203-1222, 2023. [Link]
Pruning Long Chain-of-Thought of Large Reasoning Models via Small-Scale Preference Optimization
Bin Hong, Jiayu Liu, Kai Zhang, Jianwen Sun, Mengdi Zhang, Zhenya Huang.
ICLR'2026, accepted. [PDF]
Advancing Tool-Augmented Large Language Models via Meta-Verification and Reflection Learning
Zhiyuan Ma, Jiayu Liu, Xianzhen Luo, Zhenya Huang, Qingfu Zhu, Wanxiang Che.
KDD'2025, 2078-2089, Toronto, Canada, August 3-7, 2025. [PDF] [Code]
Automated Creation of Reusable and Diverse Toolsets for Enhancing LLM Reasoning
Zhiyuan Ma, Zhenya Huang, Jiayu Liu, Minmao Wang, Hongke Zhao, Xin Li.
AAAI'2025 (Oral), 24821-24830, Philadelphia, PA, USA, February 25–March 4, 2025. [PDF] [Code]
Solving Intricate Problems with Human-like Decomposition and Rethinking
Shangzi Xue, Zhenya Huang, Jiayu Liu, Xin Lin, Yuting Ning, Binbin Jin, Xin Li, Qi Liu.
NeurIPS'2024 (Oral), 357-385, Vancouver, Canada, December 10-15, 2024. [PDF] [Code]
Learning to Solve Geometry Problems via Simulating Human Dual-Reasoning Process
Tong Xiao, Jiayu Liu, Zhenya Huang, Jinze Wu, Jing Sha, Shijin Wang, Enhong Chen.
IJCAI'2024, 6559-6568, Jeju, August 03-09, 2024. [PDF] [Code]
Bit-mask Robust Contrastive Knowledge Distillation for Unsupervised Semantic Hashing
Liyang He, Zhenya Huang, Jiayu Liu, Enhong Chen, Fei Wang, Jing Sha, Shijin Wang.
WWW'2024, 1395-1406, Singapore, May 13-17, 2024. [PDF] [Code]
Towards the Identifiability and Explainability for Personalized Learner Modeling: An Inductive Paradigm
Jiatong Li, Qi Liu, Fei Wang, Jiayu Liu, Zhenya Huang, Fangzhou Yao, Linbo Zhu, Yu Su.
WWW'2024, 3420-3431, Singapore, May 13-17, 2024. [PDF] [Code]
Enhancing the Completeness of Rationales for Multi-Step Question Answering
Shangzi Xue, Zhenya Huang, Xin Lin, Jiayu Liu, Longhu Qin, Tianhuang Su, Haifeng Liu, Qi Liu.
CIKM'2024, 2753-2763, Boise, Idaho, USA, October 21-25, 2024. [PDF] [Code]
Learning Behavior-oriented Knowledge Tracing
Bihan Xu, Zhenya Huang*, Jiayu Liu, Shuanghong Shen, Qi Liu, Enhong Chen, Jinze Wu, Shijin Wang.
KDD'2023, 2789-2800, Long Beach, CA, USA, August 6-10, 2023. [PDF] [Code]
Incremental Cognitive Diagnosis for Intelligent Education
Shiwei Tong, Jiayu Liu, Yuting Hong, Zhenya Huang, Le Wu, Qi Liu, Wei Huang, Enhong Chen, Dan Zhang.
KDD'2022, 1760-1770, Washington, DC, USA, August 14–18, 2022. [PDF]
A Mathematical Word Problem Generator with Structure Planning and Knowledge Enhancement
Longhu Qin, Jiayu Liu, Zhenya Huang*, Kai Zhang, Qi Liu, Binbin Jin, Enhong Chen.
SIGIR'2023, 1780-1784, Taipei, China, July 23-27, 2023. [PDF] [Code]
Understanding and Improving Fairness in Cognitive Diagnosis
Zheng Zhang, Le Wu, Qi Liu, Jiayu Liu, Zhenya Huang, Yu Yin, Yan Zhuang, Weibo Gao, Enhong Chen.
SCIENCE CHINA Information Sciences (SCIS), 2024, 67(5): 152106.

Services

Honors & Awards

Related Links