jdjdodood

Author: wang-jingyi

_pages/publications.md

@@ -51,6 +51,8 @@ permalink: /publications/
 
 [WWW, CCF A] Xinyao Xu, Ziyu Mao, Jianzhong Su, Xingwei Lin, David Basin, Jun Sun and Jingyi Wang\*. *Quantitative Runtime Monitoring of Ethereum Transaction Attacks*. The Web Conference, Sydney, Australia, Apr 2025. (409/2062, acceptance rate: 19.8%)
 
+[ASE, CCF A] Jianan Ma, Jingyi Wang\*, Qi Xuan and Zhen Wang. *Provable Repair of Deep Neural Network Defects through Pre-image Synthesis and Property Refinement*. The 40th IEEE/ACM International Conference on Automated Software Engineering, Seoul, South Korea, Nov 2025.
+
 [ASE, CCF A] Ziyu Mao, Xiaolin Ma, Lin Huang, Huan Yang, Wu Zhang, Weichao Sun, Yongtao Wang, Jingling Xue and Jingyi Wang\*. *Securing Millions of Decentralized Identities in Alipay Super App with End-to-End Formal Verification*. Industry Showcase Track of 40th IEEE/ACM International Conference on Automated Software Engineering, Seoul, South Korea, Nov 2025.
 
 [ISSTA, CCF A] Xiaohan Yuan, Jinfeng Li, Dongxia Wang, Yuefeng Chen, Xiaofeng Mao, Longtao Huang, Hui Xue, Wenhai Wang, Kui Ren and Jingyi Wang. *S-Eval: Automatic and Adaptive Test Generation for Benchmarking Safety Evaluation of Large Lanuguage Models.* 34th International Symposium on Software Testing and Analysis, Trondheim, Norway, Jun 2025. (107/550, acceptance rate: 19.4%, [Paper link](https://www.arxiv.org/abs/2405.14191), [Github link](https://github.com/IS2Lab/S-Eval), [HuggingFace Leaderboard Link](https://huggingface.co/spaces/IS2Lab/S-Eval))

_pages/research.md

@@ -20,12 +20,12 @@ permalink: /research/
 <!-- ### ✅ Deep Learning System Security -->
 <br>
 
-#### ***Track 1: Software Engineering for Trustworthy AI***
+#### ***Track 1: Software Engineering for Trustworthy AI Systems***
 <!-- **[TOSEM 22, ICSE 21, TACAS 21, ISSTA 21, ASE 20, ICECCS 20, ICSE 19]: Testing, Verifying and Enhancing the Robustness of Deep Learning Models** -->
 
 *We mean safe like nuclear safety as opposed to safe as in ‘trust and safety' - Ilya Sutskever*
 
-AI systems, including emerging AI models (e.g., deep neural networks and large language models), AI-based control systems (e.g., self-driving cars, robots, autonomous systems, etc) and AI-based applications (e.g., AI Chatbots, LLM agents, etc), are mostly built upon software, making it vital to ensure their trustworthiness from a software engineering perspective. In this line of research, we are working towards *a systematic testing, verification and repair framework* to evaluate, identify and fix the risks hidden in the AI models or AI-empowered systems, from different dimensions such as robustness, fairness, copyright and safety. This is crucial for stakeholders and AI-empowered industries to be aware of, manage and mitigate the risks in the new AI era.  
+AI systems, including emerging AI models (e.g., deep neural networks and large language models), AI-based cyber-physical systems (e.g., self-driving cars, robots, autonomous systems, etc) and AI-based applications (e.g., AI Chatbots, LLM agents, etc), are mostly built upon software, making it vital to ensure their trustworthiness from a software engineering perspective. In this line of research, we are working towards *a systematic testing, verification and repair framework* to evaluate, identify and fix the risks hidden in the AI models or AI-empowered systems, from different dimensions such as robustness, fairness, copyright and safety. This is crucial for stakeholders and AI-empowered industries to be aware of, manage and mitigate the risks in the new AI era.  
 
 <!-- including novel testing metrics correlated to robustness, test case generation methods, automatic verification and repair techniques to comprehensively test, verify and enhance the robustness of deep learning models deployed in various application scenarios, e.g., image classification, object detection and NLP. -->