- Human-Written vs. AI-Generated Code: A Large-Scale Study of Defects, Vulnerabilities, and Complexity
- Detecting Stealthy Data Poisoning Attacks in AI Code Generators
- Quality In, Quality Out: Investigating Training Data's Role in AI Code Generation
- Enhancing Robustness of AI Offensive Code Generators via Data Augmentation
- Enhancing AI-based Generation of Software Exploits with Contextual Information
- Conference Talks:
- Paper Presentation at ISSRE 2025
- Paper Presentation at ReSAISE 2025
- Paper Presentation at ISSRE 2024
- Paper Presentation at ICPC 2024
- Paper Presentation at ReSAISE 2023
Research Experience
- Research Project: Focuses on the trustworthiness of AI-based code generators, particularly their ability to generate functionally correct and secure source code from natural language descriptions. Aims to address limitations such as generating incorrect or insecure code due to ambiguous descriptions or training data poisoning, and to develop effective mitigation strategies.
Education
- Degree: MSc in Computer Engineering
- University: University of Naples Federico II, Italy
- Time: Not provided
- Major: Computer Engineering
- Degree: BSc in Computer Engineering
- Degree: Diploma Apple Developer Academy
- University: University of Naples Federico II in partnership with Apple, Naples, Italy
Background
- Research Interests: AI Code Generation, Software Security, Code Quality
- Professional Field: Information Technology and Electrical Engineering
- Bio: PhD candidate in Information Technology and Electrical Engineering (ITEE) at the University of Naples Federico II. Her research focuses on the security and robustness of AI-based code generators and the application of LLMs for offensive security.
Miscellany
Contact: cristina.improta@unina.it or visit DESSERT lab, Via Claudio 21, Naples, IT, Building 3/A, 4th floor.