Greetings! 👋🏻 Welcome to my homepage.

👨🏻‍🎓 About me

I am a final-year Ph.D. candidate in the AI Thrust, at The Hong Kong University of Science and Technology, Guangzhou, advised by Prof. Hao LIU and Prof. Hai YANG. I also work closely with Prof. Hui XIONG, Chair Professor, on large language models and foundation models for urban general intelligence. Before that, I received my M.Eng. in Computer Technology from Central South University in 2022. My research centers on Large Language Models and Agents, AI4Science, and spatio-temporal data mining, with a unifying goal of building agentic and ML systems for high-stakes real-world decision-making across urban, transportation, and scientific domains.

My work has moved from scientific forecasting, through automated machine learning, to today’s LLM decision-making agents, held together by one aim: building AI that is trustworthy, efficient, and genuinely deployable. It ships beyond benchmarks: ProfiLLM, among the first tool-augmented agentic systems for ride-hailing dispatch, runs on DiDi’s international platform in Brazil, serving millions of daily orders, with verified live A/B-test gains; and NRFormer is deployed online for nationwide nuclear-radiation forecasting up to 24 days ahead. This work rests on 5 first/co-author CCF-A papers (KDD, IJCAI, SIGIR), 2 first-author IEEE Transactions papers (TITS, TCBB) and 580+ citations.

📢 I am on the job market, actively seeking both academic and industry positions. Feel free to reach out!

🔥 News

• 2026.06 🎉 Delighted to receive the Best Research Award (First place) in AI at HKUSTGZ.
• 2026.05 🚕 Glad to release ProfiLLM, a utility-aligned agentic LLM framework for industrial ride-hailing dispatching, deployed on DiDi’s platform in Brazil. [project page]
• 2026.05 🎉 LLM-ODDR, an LLM framework for joint order dispatching and driver repositioning, was accepted to IEEE TITS.
• 2025.11 🎉 TelePiT was accepted by KDD 2026, focusing on global subseasonal-to-seasonal forecasting.
• 2025.09 🥳 One paper was accepted to NeurIPS 2025 about foundation models for scientific discovery.
• 2025.07 🏅 Glad to receive the KDD Student Travel Award.
• 2025.05 🌏 Thrilled to release TelePiT, a teleconnection-aware transformer for global S2S Forecasting.
• 2025.05 🍀 Glad to release LLM-ODDR, an LLM framework for joint order dispatching and driver repositioning in ride-hailing services.
• 2025.05 🎉 NRFormer was accepted to KDD 2025 about Nationwide Nuclear Radiation Forecasting.
• 2024.11 🎉 AutoSTF was accepted to KDD 2025 about Automated Spatio-Temporal Forecasting.

📑 Selected Publications

• [9]. ProfiLLM: Utility-Aligned Agentic User Profiling for Industrial Ride-Hailing Dispatch
  Tengfei Lyu*, Zirui Yuan*, Xu Liu, Kai Wan, Zihao Lu, and Hao Liu.
  VLDB SDS Track, Under Review.
  * Equal contribution. Work done during internship at Didichuxing Co. Ltd.
  [project] [code]

• [8]. Atmospheric Diffusion-Guided Spatio-Temporal Transformer for Nuclear Radiation Forecasting
  Tengfei Lyu, Jindong Han, and Hao Liu.
  TKDE, Under Review.
  [arXiv] [code]

• [7]. LLM-ODDR: A Large Language Model Framework for Joint Order Dispatching and Driver Repositioning
  Tengfei Lyu, Siyuan Feng, Hao Liu, and Hai Yang.
  IEEE Transactions on Intelligent Transportation Systems (IEEE TITS). Accepted.
  [paper] [arXiv] [code]

• [6]. Physics-Informed Teleconnection-Aware Transformer for Global Subseasonal-to-Seasonal Forecasting
  Tengfei Lyu, Weijia Zhang, and Hao Liu.
  The 32nd ACM SIGKDD Conference on Knowledge Discovery and Data Mining (KDD 2026, CCF A).
  [paper] [arXiv] [code]

• [5]. AutoSTF: Decoupled Neural Architecture Search for Cost-Effective Automated Spatio-Temporal Forecasting
  Tengfei Lyu, Weijia Zhang, Jinliang Deng, and Hao Liu.
  The 31st ACM SIGKDD Conference on Knowledge Discovery and Data Mining (KDD 2025, CCF A).
  [paper] [arXiv] [code] [Poster]

• [4]. NRFormer: Nationwide Nuclear Radiation Forecasting with Spatio-Temporal Transformer
  Tengfei Lyu, Jindong Han, and Hao Liu.
  The 31st ACM SIGKDD Conference on Knowledge Discovery and Data Mining (KDD 2025, CCF A).
  [paper] [arXiv] [code] [Poster]

• [3]. MDNN: A Multimodal Deep Neural Network for Predicting Drug-Drug Interaction Events
  Tengfei Lyu, Jianliang Gao, Ling Tian, Zhao Li, Peng Zhang and Ji Zhang.
  The 30th International Joint Conference on Artificial Intelligence (IJCAI 2021, CCF A).
  [paper] [poster] [slides]

• [2]. MGNN: A Multimodal Graph Neural Network for Predicting the Survival of Cancer Patients
  Jianliang Gao, Tengfei Lyu, Fan Xiong, Jianxin Wang, Weimao Ke and Zhao Li. (First Student Author)
  The 43rd International ACM SIGIR Conference on Research and Development in Information Retrieval (SIGIR 2020, CCF A).
  [paper] [slides]

• [1]. Predicting the Survival of Cancer Patients with Multimodal Graph Neural Network
  Jianliang Gao, Tengfei Lyu, Fan Xiong, Jianxin Wang, Weimao Ke and Zhao Li. (First Student Author)
  IEEE-ACM Transactions on Computational Biology and Bioinformatics (IEEE/ACM TCBB 2021, CCF B, JCR Q1).
  [paper]

🧑🏻‍💻 Services

Reviewer:

• SIGKDD 2024, 2025, 2026
• WebConf 2025, 2026
• NeurIPS 2023, 2024
• TKDE, TKDD, TIST, TITS

🏆 Selected Honours

• Best Research Award
  2026, AI at HKUSTGZ.

• KDD Student Travel Award
  2025, The ACM SIGKDD.

• Outstanding Paper Award
  2024, Tsinghua Science and Technology.

• Outstanding Graduate of Hunan Province (Top 1%)
  2022, Department of Education of Hunan Province.

• President Innovation Scholarship (Top 1%)
  2021, Central South University.

• National Scholarship (Top 1%)
  2020, Ministry of Education of China.

• ACM SIGIR Student Travel Award
  2020, ACM SIGIR Conference on Information Retrieval.

• National Scholarship (Top 1%)
  2018, Ministry of Education of China.

• Presidential Scholarship (Top 0.1%)
  2018, Top Undergraduate Scholarship, 10 students in University.

• Liaoning Provincial Government Scholarship (Top 1%)
  2017, Department of Education of Liaoning Province.

• National Scholarship (Top 1%)
  2016, Ministry of Education of China.