CV

Profile

I am an InnoCORE Postdoctoral Researcher in Computer Science and Engineering at UNIST, supervised by Prof. Mijung Kim. I received my Ph.D. from Nanjing University in June 2026 under the supervision of Prof. Zhenyu Chen and Prof. Chunrong Fang. My current research primarily focuses on Quality Assurance for AI Infrastructure.

Before that, I received my master’s degree from the College of Intelligence and Computing at Tianjin University, where I was supervised by Prof. Zan Wang, Prof. Shuang Liu, and Prof. Junjie Chen. I received my bachelor’s degree from Nantong University under the supervision of Prof. Xiang Chen.

Contact and Links

• Email: 602022320006@smail.nju.edu.cn
• Affiliation: UNIST
• Google Scholar: Google Scholar
• ORCID: 0000-0003-1816-2246

Education and Experience

• InnoCORE Project Postdoctoral Researcher, Department of Computer Science and Engineering, UNIST, 2026.06-2026.11
  • Supervisor: Prof. Mijung Kim
• Ph.D. in Software Engineering, Nanjing University, 2022.09-2026.06
  • Supervisors: Prof. Zhenyu Chen, Prof. Chunrong Fang
• Master’s degree, College of Intelligence and Computing, Tianjin University, 2019.09-2022.01
  • Supervisors: Prof. Zan Wang, Prof. Shuang Liu, Prof. Junjie Chen
• Bachelor’s degree, Nantong University, 2015.09-2019.06
  • Supervisor: Prof. Xiang Chen

Research and Technical Areas

• Quality assurance for AI infrastructure
• Software testing
• SE for AI
• AI for SE

Project Experience

• MindSpore Model Generalization Testing Technology, completed, student leader
  • Developed a model structure generalization tool based on the domestic deep learning framework MindSpore.
  • Extracted feature factors, ranges, and constraints of mainstream CV and NLP model architectures.
  • Modified model structures and performed testing to detect defects.
  • The project results were an essential part in supporting the collaboration team to win the Huawei 2023 Innovation Testing Application Award.
• Research on Key Mutation Testing Techniques for DL Frameworks, ongoing, student leader
  • Designed mutation rules that simulate actual user development operations to generate test models aligned with practical scenarios.
  • Evaluated the diversity and sufficiency of test data by integrating static and dynamic framework features.
  • Combined heuristic methods and experience replay to improve testing efficiency and quality.
  • Project results were published at ASE 2024, and a patent application for an invention is underway.
• Fast Detection, Localization, and Patch Generation for Concurrency Defects, completed, project participant
  • Designed a heuristic scheduling testing method for Java concurrent programs.
  • Improved detection efficiency and effectiveness.
  • Project results were published at QRS 2021 and granted an invention patent.
• LLM Generalization Testing Tool for MindSpore, completed, student leader
  • Developed a component-based generalization testing tool for large-scale models on MindSpore.
  • Enabled automated mutation and recombination of model structures such as layers, shapes, and parameters under defined constraints.
  • Built cross-framework validation pipelines by generating equivalent training scripts in PyTorch and conducting comparative analysis to detect training failures and accuracy anomalies.
  • Designed iterative strategies to identify fault-inducing factors and support multi-dimensional issue diagnosis across functionality, performance, and numerical accuracy.

Academic Service

Journal Reviewer

• ACM Computing Surveys (CSUR)
• IEEE Transactions on Dependable and Secure Computing (TDSC)

Conference Service / Reviewing

• 2026: Shadow PC, ICSE
• 2026: Co-reviewer, ISSTA
• 2026: Co-reviewer, ICSE
• 2025: Co-reviewer, PROMISE
• 2025: Co-reviewer, FSE

Other Reviewing

• Co-reviewer, Journal of Software

Awards and Honors

• Huawei 2023 Innovation Testing Application Award: project results from MindSpore Model Generalization Testing Technology supported the collaboration team in winning this award.
• Outstanding Graduate of Tianjin University — 2022.06
• Outstanding Communist Youth League Member of Tianjin University — 2021.01
• Outstanding Undergraduate Graduation Project/Thesis of Nantong University, Class of 2019 — 2019.06
• Outstanding Student Cadre of Jiangsu Province — 2018.05
• National Scholarship for Undergraduate Students — 2017.11
• Merit Student of Nantong City — 2017.05
• Model Merit Student of Nantong University — 2016.09

Publications

CIPIHunter: Detecting Configuration-Induced Prediction Instability in Deep Learning Frameworks

Yanzhou Mu, Shuo Meng, Mijung Kim, Xiang Chen, Chunrong Fang, Zhenyu Chen, Juan Zhai. CIPIHunter: Detecting Configuration-Induced Prediction Instability in Deep Learning Frameworks. In Proceedings of IEEE/ACM Automated Software Engineering Conference (ASE 2026). ACM, Munich, Germany. CCF-A.

Understanding LLM-Centric Challenges for Deep Learning Frameworks: An Empirical Analysis

Yanzhou Mu, Rong Wang, Juan Zhai, Chunrong Fang, Xiang Chen, Jiacong Wu, An Guo, Jiawei Shen, Bingzhuo Li, and Zhenyu Chen. Understanding LLM-Centric Challenges for Deep Learning Frameworks: An Empirical Analysis. ACM Transactions on Software Engineering and Methodology, 2026. CCF-A.

Deep Learning Framework Testing via Model Mutation: How far are we?

Yanzhou Mu, Rong Wang, Juan Zhai, Chunrong Fang, Xiang Chen, Peiran Yang, Zhixiang Cao, Ruixiang Qian, Shaoyu Yang, and Zhenyu Chen. Deep Learning Framework Testing via Model Mutation: How far are we? IEEE Transactions on Software Engineering, 2025. CCF-A.

Improving Deep Learning Framework Testing with Model-Level Metamorphic Testing

Yanzhou Mu, Juan Zhai, Chunrong Fang, Xiang Chen, Zhixiang Cao, Peiran Yang, Kexin Zhao, An Guo, and Zhenyu Chen. Improving Deep Learning Framework Testing with Model-Level Metamorphic Testing. International Symposium on Software Testing and Analysis (ISSTA 2025), Thondheim, Norway, 2025. CCF-A.

DevMuT: Testing Deep Learning Framework via Developer Expertise-Based Mutation

Yanzhou Mu, Juan Zhai, Chunrong Fang, Xiang Chen, Zhixiang Cao, Peiran Yang, Yinglong Zou, Tao Zheng, and Zhenyu Chen. DevMuT: Testing Deep Learning Framework via Developer Expertise-Based Mutation. In Proceedings of IEEE/ACM Automated Software Engineering Conference (ASE 2024). ACM, Sacramento, California, United States. CCF-A.

采用多目标优化的深度学习测试优化方法 / Deep learning test optimization method using multi-objective optimization

Chinese version: 沐燕舟, 王赞, 陈翔, 陈俊洁, 赵静珂, 王建敏. 采用多目标优化的深度学习测试优化方法[J]. 软件学报, 2022, 33(07):2499-2524. DOI:10.13328/j.cnki.jos.006583. English version: Yanzhou Mu, Zan Wang, Xiang Chen, Junjie Chen, Jingke Zhao, and Jianmin Wang. Deep learning test optimization method using multi-objective optimization. International Journal of Software & Informatics, 12(4), 2022.

HARS: Heuristic-enhanced adaptive randomized scheduling for concurrency testing

Yanzhou Mu, Zan Wang, Shuang Liu, Jun Sun, Junjie Chen, and Xiang Chen. HARS: Heuristic-enhanced adaptive randomized scheduling for concurrency testing. In 2021 IEEE 21st International Conference on Software Quality, Reliability and Security (QRS), pages 219-230. IEEE, 2021.

Can test input selection methods for deep neural network guarantee test diversity?

Chunyu Zhao, Yanzhou Mu, Xiang Chen, Jingke Zhao, Xiaolin Ju, and Gan Wang. Can test input selection methods for deep neural network guarantee test diversity? A large-scale empirical study. Information and Software Technology, 150:106982, 2022. Co-first author

Revisiting heterogeneous defect prediction methods: How far are we?

Xiang Chen, Yanzhou Mu, Ke Liu, Zhanqi Cui, and Chao Ni. Revisiting heterogeneous defect prediction methods: How far are we? Information and Software Technology, 130:106441, 2021. Co-first author

Do different cross-project defect prediction methods identify the same defective modules?

Xiang Chen, Yanzhou Mu, Yubin Qu, Chao Ni, Meng Liu, Tong He, and Shangqing Liu. Do different cross-project defect prediction methods identify the same defective modules? Journal of Software: Evolution and Process, 32(5):e2234, 2020. Co-first author

Train in Vain: Functionality-Preserving Poisoning to Prevent Unauthorized Use of Code Datasets

Yuan Xiao, Yuchen Chen, Jiaming Wang, Wei Song, Jun Sun, Shiqing Ma, Yanzhou Mu, Juan Zhai, Chunrong Fang, Jin Song Dong, and Zhenyu Chen. Train in Vain: Functionality-Preserving Poisoning to Prevent Unauthorized Use of Code Datasets. Findings of the Association for Computational Linguistics, 2026. CCF A.

When Autonomous Vehicle Meets V2X Cooperative Perception: How Far Are We?

An Guo, Shuoxiao Zhang, Enyi Tang, Xinyu Gao, Haomin Pang, Haoxiang Tian, Yanzhou Mu, Wu Wen, Chunrong Fang, and Zhenyu Chen. When Autonomous Vehicle Meets V2X Cooperative Perception: How Far Are We? The 40th IEEE/ACM International Conference on Automated Software Engineering, November 16-20, 2025, Seoul, South Korea. CCF-A.

Deep Learning Framework Testing via Heuristic Guidance Based on Multiple Model Measurements

Yilong Zou, Juan Zhai, Chunrong Fang, Yanzhou Mu, Jiawei Liu, and Zhenyu Chen. Deep Learning Framework Testing via Heuristic Guidance Based on Multiple Model Measurements. ACM Transactions on Software Engineering and Methodology, 2026. CCF-A.

An Empirical Study of Bugs in Modern LLM Agent Frameworks Permalink

Xinxue Zhu, Jiacong Wu, Xiaoyu Zhang, Tianlin Li, Yanzhou Mu, Juan Zhai, Chao Shen, Chunrong Fang, and Yang Liu. An Empirical Study of Bugs in Modern LLM Agent Frameworks. The 2nd International Workshop on Large Language Model Supply Chain Analysis (LLMSC 2026), co-located with FSE 2026, 2026. Accepted. DOI:10.48550/arXiv.2602.21806. Workshop arXiv | PDF | DOI | Workshop

Cross-project defect prediction via asttoken2vec and blstm-based neural network

Hao Li, Xiaohong Li, Xiang Chen, Xiaofei Xie, Yanzhou Mu, and Zhiyong Feng. Cross-project defect prediction via asttoken2vec and blstm-based neural network. In 2019 International Joint Conference on Neural Networks (IJCNN), pages 1-8. IEEE, 2019.

Generate Realistic Test Scenes for V2X Communication Systems

An Guo, Xinyu Gao, Chunrong Fang, Haoxiang Tian, Weisong Sun, Yanzhou Mu, Shuncheng Tang, Lei Ma, Xiapu Luo, and Zhenyu Chen. Generate Realistic Test Scenes for V2X Communication Systems. ACM Transactions on Software Engineering and Methodology, 2025. CCF-A. Major revision.

Tactile-based Multimodal Fusion in Embodied Intelligence: A Survey of Vision, Language, and Contact-Driven Paradigms Permalink

Zhixiang Cao, Di Tian, Runwei Guan, Yanzhou Mu, Xiaolou Sun, Shaofeng Liang, Daizong Liu, Tao Huang, Yutao Yue, Henghui Ding, Bin Fang, Alex Zhou, Qing-Long Han, and Hui Xiong. Tactile-based Multimodal Fusion in Embodied Intelligence: A Survey of Vision, Language, and Contact-Driven Paradigms. arXiv preprint, under review, 2026. DOI:10.48550/arXiv.2605.17336. Preprint arXiv | PDF | DOI

Patents

• Yanzhou Mu; Zan Wang; Shuang Liu. A heuristic rule-based concurrent adaptive random testing method. Patent No: CN113468047A.
• Junjie Chen; Yanzhou Mu; Zan Wang; Jianmin Wang; Jiao Jia. A multi-objective optimization-based method for selecting deep learning test inputs. Patent No: CN114721934A.
• Zhenyu Chen; Yidong Ke; Jiawei Liu; Yanzhou Mu. A neural network fuzz testing method guided by mutation image entropy values. Patent No: CN117218051A.
• Yongping Liu; Zhenyu Chen; Chunrong Fang; Yanzhou Mu; Ruifa Luo; Shuiyong Du. Mutation testing method, device, and computer equipment based on neuron characteristics of intelligent traffic models. Patent No: CN118689772A.
• Yongping Liu; Zhenyu Chen; Chunrong Fang; Yanzhou Mu; Ruifa Luo; Shuiyong Du. Domain-specific mutation testing method, device, and computer equipment based on intelligent traffic V2X models. Patent No: CN118689771A.
• Zhijin Guan; Haiying Ma; Hehe Gu; Zongyuan Zhang; Yanzhou Mu; Jing Zhao; Meng Liu. A quantum circuit image recognition method. Patent No: CN106997462A.