Zheye Deng 邓哲也
About Me
Hi! I am Zheye Deng (邓哲也). I am a fourth-year Ph.D. student at The Hong Kong University of Science and Technology advised by Prof. Yangqiu Song. I received my B.Sc. in Computer Science from Peking University, where I was advised by Prof. Ming Zhang.
I have worked on structured knowledge and reasoning, from constructing large-scale knowledge graphs to extracting and integrating structural information from text. My current research interests lie in LLM post-training and agentic reinforcement learning, where I develop learning agents that combine structured reasoning and tool use to tackle complex decision-making and problem-solving tasks.
Education
The Hong Kong University of Science and Technology · Ph.D. in Computer ScienceSept. 2022 – Now
Peking University · B.Sc. in Computer ScienceSept. 2016 – Jun. 2020
Experience
Amazon Stores Foundational AI, Amazon · Applied Scientist Intern · Palo Alto, CAMay 2026 – Aug. 2026
Alimama (Search & Advertising), Alibaba Group · Research Scientist Intern · BeijingJan. 2026 – Apr. 2026
Amazon Stores Foundational AI, Amazon · Applied Scientist Intern · Palo Alto, CAJun. 2025 – Dec. 2025
Megvii Infrastructure Team, Megvii · R&D Engineer · BeijingJul. 2020 – Aug. 2022
UCSB NLP Group, UC Santa Barbara · Research Assistant · Santa Barbara, CAJun. 2019 – Sept. 2019
Douyin Data Team, ByteDance · R&D Engineer Intern · BeijingJun. 2018 – Oct. 2018
Awards
- EMNLP 2024 Outstanding Paper Award (2024)
- Hong Kong PhD Fellowship (2022 - 2026)
- HKUST RedBird Academic Excellence Award for Continuing PhD Students (2024 - 2025)
- HKUST RedBird Academic Excellence Award for Continuing PhD Students (2023 - 2024)
- HKUST RedBird PhD Scholarship (2022)
- Peking University Third-class Scholarship (2019)
- Third Runner-up in WSDM Cup (2018)
- Second Prize in Peking University Programming Contest (2017)
- Gold Prize in Asia-Pacific Informatics Olympiad (2015)
- Silver Prize in National Olympiad in Informatics (2015)
Contact Me
You can reach me at zdengah [at] cse [dot] ust [dot] hk.