Jingheng Ye is a second-year Ph.D. student in Computer Engineering at Northeastern University (NEU), advised by Prof. Weiyan Shi at the CHATS Lab. Prior to NEU, he obtained his M.S. in Computer Science from Tsinghua University in 2025, where he was a member of the THUKE Lab advised by Prof. Hai-Tao Zheng. He received his B.S. from South China University of Technology in 2022.

His research centers on recursive self-improving AI and AI safety. His current interests include:

• Recursive self-improving AI: Optimize agent harness for higher performance and lower cost.
• Agent Safety & Evaluation: Detecting and benchmarking unsafe behaviors in LLM agents, specifically assessing the risks posed to Human-AI Collaboration in high-stakes scenarios.

You can find his CV here: Jingheng Ye’s CV.

🔥 News

• 2026.09

  🎉 Coding with the Enemy accepted to NeurIPS 2026!

• 2026.07

  🏆 Coding with the Enemy won the Best Paper Award at the DL4C Workshop @ ICML 2026.

• 2026.05

  🎉 Three papers accepted to ACL 2026 (two Main, one Findings).

• 2025.09

  🎓 Started my Ph.D. journey at Northeastern University!

• 2025.09

  🎉 Two papers accepted to EMNLP 2025 (one Industry Track, one Main Conference).

More news

• 2025.05

  🎉 One paper accepted to ACL 2025 (CLEME 2.0).

• 2024.12

  🎉 EXCGEC is accepted by AAAI 2025 (Oral).

• 2024.12

  🏅 Awarded the National Scholarship.

Selected Publications

• Coding with the Enemy: Can Human Developers Detect AI Agent Sabotage?

  Jingheng Ye*, Huiqi Zou*, Simon Yu, Weiyan Shi
  Best Paper Award, DL4C @ ICML 2026
  NeurIPS 2026 Paper / Code / Project

• CL²GEC: A Multi-Discipline Benchmark for Continual Learning in Chinese Literature Grammatical Error Correction

  Shang Qin, Jingheng Ye, Yinghui Li, et al.
  ACL 2026 Paper

• Position: LLMs Can be Good Tutors in English Education

  Jingheng Ye, Shen Wang, Deqing Zou, et al.
  EMNLP 2025 Paper

• ProductAgent: Benchmarking Conversational Product Search Agent with Asking Clarification Questions

  Jingheng Ye, Yong Jiang, Xiaobin Wang, et al.
  EMNLP 2025 Industry Paper

• EXCGEC: A Benchmark for Edit-Wise Explainable Chinese Grammatical Error Correction

  Jingheng Ye*, Shang Qin*, Yinghui Li, et al.
  AAAI 2025 Oral Paper / Code

• CLEME2.0: Towards Interpretable Evaluation by Disentangling Edits for Grammatical Error Correction

  Jingheng Ye*, Zishan Xu*, Yinghui Li, et al.
  ACL 2025 Paper / Code

• CLEME: Debiasing Multi-reference Evaluation for Grammatical Error Correction

  Jingheng Ye*, Yinghui Li*, Qingyu Zhou, et al.
  EMNLP 2023 Paper / Code

• MixEdit: Revisiting Data Augmentation and Beyond for Grammatical Error Correction

  Jingheng Ye, Yinghui Li, Yangning Li, et al.
  EMNLP 2023 Findings Paper / Code

• A Frustratingly Easy Plug-and-Play Detection-and-Reasoning Module for Chinese Spelling Check

  Haojing Huang*, Jingheng Ye*, Qingyu Zhou, et al.
  EMNLP 2023 Findings Paper / Code

* Equal contribution. Full list on Google Scholar.

Experience

• Northeastern University, Boston, MA Sep 2025 - Present
  Research Assistant

  Advisor: Prof. Weiyan Shi. Focus: Code Sabotage & Agent Safety.

• Squirrel AI, Shanghai, China Aug 2024 - Jun 2025
  Research Intern

  Advisor: Dr. Qingsong Wen. Topic: AI Education.

• Alibaba (Tongyi Lab), Hangzhou, China Dec 2023 - Aug 2024
  Research Intern

  Advisor: Dr. Yong Jiang. Topic: LLM Agents.

• Tencent, Shenzhen, China Jul 2021 - Sep 2021
  Software Engineering Intern

  Topic: Database System Testing (TDSQL).

Honors and Awards

• 2026.07

  Best Paper Award, DL4C Workshop @ ICML 2026

• 2025.06

  Outstanding Graduate of Beijing (Top 1%)

• 2024.12

  National Scholarship (Top 0.2%)

• 2023.08

  CCL23-Eval CLTC Track-1, Second Prize

• 2022.10

  CCL22-Eval CLTC Task-4, First Prize

• 2019.12

  National Scholarship (Top 0.2%)