Zhe Wang
Logo Graduate Student, University of Illinois at Urbana-Champaign
Logo Undergraduate Student, Tsinghua University

Hi, there! I'm Zhe Wang, a first year M.S. student at University of Illinois at Urbana-Champaign, majoring in Computer Science. I am advised by Prof. Lingming Zhang. Before coming to UIUC, I completed my undergraduate study at Tsinghua University, majoring in Mathematics and Physics.

Research Interests

My research interest lies in the intersection of Artificial Intelligence and Software Engineering. More specifically:

  • LLMs for Code: to develop LLMs to solve software engineering tasks through post-training via synthetic data
  • Trustworthy LLMs: to enhance trustworthiness, resilience and reliability of helpful-only LLMs against vulnerable code and malicious cyberactivity attacks
  • LLM Applications: to empower LLMs with reasoning, planning and collaboration capabilities through alignment training and agent-based systems

I am currently seeking for an Ph.D. position starting from Fall 2026. Feel free to drop me an email if you are interested in my research or have any questions. Please find my detailed Curriculum Vitae below:

Curriculum Vitae

Selected Publications (view all )
PurpCode: Reasoning for Safer Code Generation
PurpCode: Reasoning for Safer Code Generation

Jiawei Liu*, Nirav Diwan*, Zhe Wang*, Haoyu Zhai, Xiaona Zhou, Kiet A. Nguyen, Tianjiao Yu, Muntasir Wahed, Yinlin Deng, Hadjer Benkraouda, Yuxiang Wei, Lingming Zhang, Ismini Lourentzou, Gang Wang (* equal contribution)

🥇 1st Place in Amazon Nova AI Challenge 2025 ($250,000)

We introduce PurpCode, the first post-training recipe for training safe code reasoning models towards generating secure code and defending against malicious cyberactivities. PurpCode trains a reasoning model in two stages: (i) Rule Learning, which explicitly teaches the model to reference cybersafety rules to generate vulnerability-free code and to avoid facilitating malicious cyberactivities; and (ii) Reinforcement Learning, which optimizes model safety and preserves model utility through diverse, multi-objective reward mechanisms.

PurpCode: Reasoning for Safer Code Generation

Jiawei Liu*, Nirav Diwan*, Zhe Wang*, Haoyu Zhai, Xiaona Zhou, Kiet A. Nguyen, Tianjiao Yu, Muntasir Wahed, Yinlin Deng, Hadjer Benkraouda, Yuxiang Wei, Lingming Zhang, Ismini Lourentzou, Gang Wang (* equal contribution)

🥇 1st Place in Amazon Nova AI Challenge 2025 ($250,000)

We introduce PurpCode, the first post-training recipe for training safe code reasoning models towards generating secure code and defending against malicious cyberactivities. PurpCode trains a reasoning model in two stages: (i) Rule Learning, which explicitly teaches the model to reference cybersafety rules to generate vulnerability-free code and to avoid facilitating malicious cyberactivities; and (ii) Reinforcement Learning, which optimizes model safety and preserves model utility through diverse, multi-objective reward mechanisms.

MultiAgentBench: Evaluating the Collaboration and Competition of LLM agents
MultiAgentBench: Evaluating the Collaboration and Competition of LLM agents

Kunlun Zhu†, Hongyi Du†, Zhaochen Hong†, Xiaocheng Yang†, Shuyi Guo†, Zhe Wang†, Zhenhailong Wang, Cheng Qian, Xiangru Tang, Heng Ji, Jiaxuan You († core contributors)

ACL 2025 Main

In this paper, we introduce MultiAgentBench, a comprehensive benchmark designed to evaluate LLM-based multi-agent systems across diverse, interactive scenarios. Our framework measures not only task completion but also the quality of collaboration and competition using novel, milestone-based key performance indicators.

MultiAgentBench: Evaluating the Collaboration and Competition of LLM agents

Kunlun Zhu†, Hongyi Du†, Zhaochen Hong†, Xiaocheng Yang†, Shuyi Guo†, Zhe Wang†, Zhenhailong Wang, Cheng Qian, Xiangru Tang, Heng Ji, Jiaxuan You († core contributors)

ACL 2025 Main

In this paper, we introduce MultiAgentBench, a comprehensive benchmark designed to evaluate LLM-based multi-agent systems across diverse, interactive scenarios. Our framework measures not only task completion but also the quality of collaboration and competition using novel, milestone-based key performance indicators.

Magicoder: Empowering Code Generation with OSS-Instruct
Magicoder: Empowering Code Generation with OSS-Instruct

Yuxiang Wei, Zhe Wang, Jiawei Liu, Yifeng Ding, Lingming Zhang

ICML 2024

In this paper, we introduce Magicoder, a series of fully open-source (code, weights, and data) Large Language Models (LLMs) for code that significantly closes the gap with top code models while having no more than 7B parameters.

Magicoder: Empowering Code Generation with OSS-Instruct

Yuxiang Wei, Zhe Wang, Jiawei Liu, Yifeng Ding, Lingming Zhang

ICML 2024

In this paper, we introduce Magicoder, a series of fully open-source (code, weights, and data) Large Language Models (LLMs) for code that significantly closes the gap with top code models while having no more than 7B parameters.

All publications
Education
  • University of Illinois at Urbana-Champaign
    University of Illinois at Urbana-Champaign
    M.S. in Computer Science
    Aug. 2024 - Present
  • Tsinghua University
    Tsinghua University
    B.S. in Mathematics and Physics
    Sep. 2020 - Jul. 2024
  • Shanghai High School
    Shanghai High School
    High School
    Sep. 2017 - Jun. 2020