CV

Mingzhe Du's Curriculum Vitae

General Information

Full Name Mingzhe Du
Phone (+65) 9658 2486
Email dumingzhex@gmail.com
Website https://mingzhe.space
Address Innovation 4.0, 3 Research Link, Singapore, 117602
Languages Mandarin (native), English (professional)

Research Interests

  • Self-evolving Harness / Long-horizon Code Generation / Preference Alignment

Education

  • Aug. 2022 - Present

    Singapore

    PhD Candidate in Computer Science
    Nanyang Technological University, College of Computing and Data Science
    • Advisor: Prof. Anh Tuan Luu
  • Oct. 2017 - Dec. 2019

    Melbourne, Australia

    Master of Information Technology
    The University of Melbourne, Faculty of Engineering
    • Advisor: Prof. Richard O. Sinnott
    • Specialization: Distributed Computing (Honorable Dissertation)
  • Sep. 2013 - Jun. 2017

    Changsha, China

    Bachelor of Information Security
    Central South University, School of Computer Science and Engineering
    • {"Excellent Scholarships (GPA"=>"3.7/4.0)"}

Professional Experience

  • Aug. 2021 - Present
    Research Associate
    National University of Singapore, Institute of Data Science
    • Advisor: Prof. See-Kiong Ng
    • Affiliated with Cisco-NUS Accelerated Digital Economy Corporate Laboratory.
  • Dec. 2019 - Aug. 2021
    Software Architect (Full Time)
    ByteDance, TikTok Search
    • Architected a scalable, general-purpose search engine (Search Eternal) that now powers core search services for major ByteDance platforms, including Douyin, TikTok and TouTiao.
  • Nov. 2018 - Mar. 2019
    Research Intern
    Microsoft, Asia-Pacific Research and Development Group
    • Mentor: Wei Liu
    • Designed and implemented a Transformer-based model for text-to-image synthesis, generating visually coherent images from textual descriptions.
  • Jul. 2016 - Oct. 2016
    Software Engineer Intern
    Alibaba, Ant Finance Core Clearing Team
    • Developed components for a core system in Alipay's payment clearing flow, enabling the aggregation of millions of orders at minute-level latency for data exchange with banking partners.
  • Mar. 2016 - Jun. 2016
    Software Engineer Intern
    Tencent, Social Network Group
    • Built and deployed a URL analysis platform to process and detect suspicious content in high-volume public message traffic, significantly improving the filtering of malicious information to enhance content security.

Academic Services

  • Conference Reviewer
    • NeurIPS (24,25,26), ICLR (24,25,26), ICML (25,26), AISTATS (24,26), AAAI (26,27), CIKM (25,26), AISI'26, ECAI'25, NAACL'24, EMNLP'24, ECAI'24, CLEF'23
  • Journal Reviewer
    • ACM Computing Survey, Frontiers in Artificial Intelligence
  • Volunteer
    • NeurIPS'25, ACL'25, WWW'24, EMNLP'24, AAAI'23
  • STEM Mentor
    • The University of Melbourne (2019-Present)

Selected Publications

  • [NeurIPS'25] Afterburner: Reinforcement Learning Facilitates Self-Improving Code Efficiency Optimization.
    Mingzhe Du, Luu Anh Tuan, Yue Liu, Yuhao Qing, Dong Huang, Xinyi He, Qian Liu, Zejun Ma, See-Kiong Ng
  • [ACL'25] CodeArena: A collective evaluation platform for LLM code generation.
    Mingzhe Du, Luu Anh Tuan, Bin Ji, Xiaobao Wu, Dong Huang, Terry Yue Zhuo, Qian Liu, See-Kiong Ng
  • [NeurIPS'24] Mercury: A Code Efficiency Benchmark for Code Large Language Models.
    Mingzhe Du, Luu Anh Tuan, Bin Ji, Qian Liu, See-Kiong Ng

Publications

  • [Preprint] Paper Espresso: From Paper Overload to Research Insight.
    Mingzhe Du, Luu Anh Tuan, Dong Huang, See-Kiong Ng.
  • [Preprint] Mastermind: Strategy-Level Learning for Repository-Scale Vulnerability Reproduction.
    Mingzhe Du, Luu Anh Tuan, Dong Huang, Renyang Liu, Tianyi Wu, Zhijiang Guo, See-Kiong Ng.
  • [Preprint] Secure Code Generation via Online Reinforcement Learning with Vulnerability Reward Model.
    Tianyi Wu, Mingzhe Du, Yue Liu, Chengran Yang, Terry Yue Zhuo, Jiaheng Zhang, See-Kiong Ng.
  • [Preprint] Scaling Code LLM Training and Test-Time Inference via Execution-Free Reward Models.
    Xiao Zhu, Xinyu Zhou, Boyu Zhu, Hanxu Hu, Mingzhe Du, Haotian Zhang, Huiming Wang, Zhijiang Guo.
  • [ICML'26] SWE-Perf: Can Language Models Optimize Code Performance on Real-World Repositories?
    Xinyi He, Qian Liu, Mingzhe Du, Lin Yan, Zhijie Fan, Yiming Huang, Zejian Yuan, Zejun Ma.
  • [ICML'26 DL4C] Nexus: Execution-Grounded Multi-Agent Test Oracle Synthesis.
    Dong Huang, Mingzhe Du+, Jie M. Zhang, Zheng Lin, Meng Luo, Qianru Zhang, See-Kiong Ng.
  • [ACL'26] Semantics-Aligned, Curriculum-Driven, and Reasoning-Enhanced Vulnerability Repair Framework (Best Paper Nomination).
    Chengran Yang, Ting Zhang, Jinfeng Jiang, Xin Zhou, Haoye Tian, Mingzhe Du, Jieke Shi, Junkai Chen, Yikun Li, Eng Lieh Ouh, Lwin Khin Shar, David Lo.
  • [TOSEM'26] Benchmarking LLMs for Unit Test Generation from Real-World Functions.
    Dong Huang, Jie M. Zhang, Mark Harman, Qianru Zhang, Mingzhe Du+, See-Kiong Ng.
  • [EACL'26] Pro-QuEST: Prompt-chaining Quiz Engine for testing Specialized Technical Product Knowledge.
    Sujatha Das Gollapalli, Mouad Hakam, Mingzhe Du, See-Kiong Ng, Mohammed Hamzeh.
  • [AMIYA'26] Improving Arabic Dialectness in LLMs with Reinforcement Learning.
    Sujatha Das Gollapalli, Mouad Hakam, Mingzhe Du, See-Kiong Ng, Mohammed Hamzeh.
  • [ICLR'26] Beyond Prompt-Induced Lies: Investigating LLM Deception on Benign Prompts. (Oral)
    Zhaomin Wu, Mingzhe Du, See-Kiong Ng, Bingsheng He.
  • [EMNLP'25] On Assigning Product and Software Codes to Service Requests with Large Language Models.
    Sujatha Das Gollapalli, Mouad Hakam, Mingzhe Du, See-Kiong Ng, Mohammed Hamzeh.
  • [NeurIPS'25] EffiBench-X: A Multi-Language Benchmark for Measuring Efficiency of LLM-Generated Code.
    Yuhao Qing*, Boyu Zhu*, Mingzhe Du*, Zhijiang Guo, Terry Yue Zhuo, Qianru Zhang, Jie M. Zhang, Heming Cui, Siu-Ming Yiu, Dong Huang, See-Kiong Ng, Luu Anh Tuan.
  • [NeurIPS'25] Guardreasoner-VL: Safeguarding VLMs via Reinforced Reasoning.
    Yue Liu, Shengfang Zhai, Mingzhe Du, Yulin Chen, Tri Cao, Hongcheng Gao, Cheng Wang, Xinfeng Li, Kun Wang, Junfeng Fang, Jiaheng Zhang, Bryan Hooi.
  • [ICSE'25] Measuring the Influence of Incorrect Code on Test Generation.
    Dong Huang, Jie M. Zhang, Mark Harman, Mingzhe Du+, Heming Cui.
  • [ACL'25] AntiLeakBench: Preventing Data Contamination by Automatically Constructing Benchmarks with Updated Real-World Knowledge. (Oral and SAC Highlight Award)
    Xiaobao Wu, Liangming Pan, Yuxi Xie, Ruiwen Zhou, Shuai Zhao, Yubo Ma, Mingzhe Du, Rui Mao, Anh Tuan Luu, William Yang Wang.
  • [ICML'25] Current Model Licensing Practices are Dragging Us into a Quagmire of Legal Noncompliance. (Oral)
    Moming Duan*, Mingzhe Du*, Rui Zhao, Mengying Wang, Yinghui Wu, Nigel Shadbolt, Bingsheng He.
  • [AAAI'25] Towards Verifiable Text Generation with Generative Agent. (Oral)
    Bin Ji, Huijun Liu, Mingzhe Du, Shasha Li, Xiaodong Liu, Jun Ma, Jie Yu, See-Kiong Ng.
  • [SAC'25] Curriculum Demonstration Selection for In-Context Learning.
    Duc Anh Vu, Cong-Duy Nguyen, Xiaobao Wu, Nhat Hoang, Mingzhe Du, Thong Nguyen, Anh Tuan Luu.
  • [JMIR'25] Unraveling Online Mental Health Through the Lens of Early Maladaptive Schemas: AI-Enabled Content Analysis of Online Mental Health Communities.
    Beng Heng Ang, Sujatha Das Gollapalli, Mingzhe Du, See-Kiong Ng.
  • [AAAI'24] From Static to Dynamic: Knowledge Metabolism for Large Language Models.
    Mingzhe Du, Anh Tuan Luu, Bin Ji, See-Kiong Ng.
  • [AAAI'24] Chain-of-Thought Improves Text Generation with Citations in Large Language Models.
    Bin Ji, Huijun Liu, Mingzhe Du, See-Kiong Ng.
  • [ECAI'24] Counseling Responses for Mental Health Forum Questions with Maladaptive Schema Prediction.
    Sujatha Das Gollapalli, Beng Heng Ang, Mingzhe Du, See-Kiong Ng.
  • [TLDK'23] Constituency-Informed and Constituency-Constrained Extractive Question Answering with Heterogeneous Graph Transformer.
    Mingzhe Du, Mouad Hakam, See-Kiong Ng, Stephane Bressan.
  • [WWW'23] Identifying Checkworthy Cure Claims on Twitter.
    Sujatha Das Gollapalli, Mingzhe Du, See-Kiong Ng.
  • [AAAI'23] Generating Reflective Questions for Engaging Gallery Visitors in ArtMuse.
    Sujatha Das Gollapalli, Mingzhe Du, See-Kiong Ng.
  • [CLEF'22] NUS-IDS at CheckThat! 2022: Identifying Check-worthiness of Tweets using CheckthaT5.
    Mingzhe Du, Sujatha Das Gollapalli, See-Kiong Ng.
  • Note: + Corresponding Author, * Equal Contribution.

Awards & Honors

  • NVIDIA Academic Grant: Predicting and Optimizing LLVM Compiler Pass Order in a Unified Model.
  • Thinking Machines Lab Research Grant: Scaling Reinforcement Learning for Efficient Code Generation.
  • NVIDIA Academic Grant: Scaling Laws of Reinforcement Learning for Code Efficiency Optimization.
  • Staff Research Award at Institute of Data Science, National University of Singapore, 2025.
  • Google Academic Research Award: SafeCodeX: Security-Aware Code Generation with LLMs, 2025.
  • First Runner-Up Award at the Singapore Healthcare AI Datathon: Identifying ICU Patient with Central Lines at Risk of CLABSI, 2024.
  • First Prize at Catalyst: A Path Recommendation System based on Data Aggregation and Analysis, 2019.
  • First Prize at Codebrew: After Dark: Finding Safe and Fast Routes for Pedestrians at Night, 2019.
  • Technology Pioneer Award at the Cybersecurity Talent Development and Entrepreneurship Forum, 2015.
  • First Prize at the Mathematical Contest in Modeling: Study on the Wreckage Position of Malaysia Airlines Flight MH370, 2014.

Miscellaneous

LinkedIn dumingzhe
GitHub elfsong
WeChat spirit_song
Last Updated 27 June, 2026