Pietro Liguori

PIETRO LIGUORI

Assistant Professor
Department of Electrical Engineering and Information Technology (DIETI)
University of Naples Federico II, Naples, Italy

Email: pietro.liguori@unina.it

Member of the DESSERT Research Group

Latest News

Updates on research, tools, and initiatives

NEW December 2025

OSS-Forge Launched on Hugging Face 🤗

In collaboration with UNC Charlotte, we launched OSS-Forge, a new open organization on Hugging Face hosting ready-to-use models, datasets, and tools for secure and reliable AI code generation.

Includes: DeVAIC (vulnerability detector), Shellcode_IA32 (NLP-to-shellcode dataset), and Human_vs_AI_Code_Quality (benchmark suite).

ACCEPTED October 2025

Large-Scale Study on Human vs. AI Code Quality

Our paper “Human-Written vs. AI-Generated Code: A Large-Scale Study of Defects, Vulnerabilities, and Complexity” has been accepted at ISSRE 2025.

Replication package available on GitHub.

OPEN Until Jan 25, 2026

CyberChallenge.IT 2026: Registrations Open!

The national cybersecurity training program for high-school and university students (ages 16–24) is now open for registrations.

As Program Coordinator for the University of Naples Federico II, I support student selection, training, and local finals. Apply now →

View all publications

Research Interests

Security, testing, and evaluation of AI-driven code generation.

Security of AI Code Generators

Studying the security properties of AI-based code generators, including data poisoning attacks, robustness to adversarial inputs, and the detection of vulnerabilities in AI-generated code. Particular attention is given to systematic differences between human-written and AI-generated code.

Neural Fault Injection

Investigating the use of AI code generators to inject realistic residual bugs into software systems, enabling large-scale fault injection testing and the evaluation of software reliability and failure modes.

Offensive Code Generation

Exploring the automatic generation of software exploits and attack code from natural language using AI models, with applications in security testing and vulnerability assessment.

Teaching Activities

Academic instruction and mentorship in cybersecurity and software engineering

Programming Workshop (9 CFU)

University of Naples Federico II, Italy
2023/24, 2024/25, 2025/26

Professor. MSc course in Python programming for Biomedical and Electronic Engineering students.

Impianti di Elaborazione (9 CFU)

University of Naples Federico II, Italy
2018/19 - Present

Teaching Assistant. MSc course in Computer Engineering on advanced topics in system performance and reliability.

CyberChallenge.IT

University of Naples Federico II, Italy
2024, 2025, 2026

Official representative and coordinator for the CyberChallenge.IT, the Italian national training program in cybersecurity.

Practical Cybersecurity (9 CFU)

Accademia Aeronautica, Italy
2025/26

Professor. Practical course on cybersecurity for the Italian Air Force Academy.

Reliability and Security of Complex Software Systems (6 CFU)

University of Campania Luigi Vanvitelli, Italy
2021/22, 2022/23

Adjunct Professor. MSc course in Computer Engineering focusing on reliability and security in complex software systems.

Supervised Theses

Bachelor’s and Master’s theses advised/co-advised (total: 30+)

Supervising students during their theses journey is one of my main motivations as an academic. Here is the full list of students I have supervised/co-supervised:

Impianti di Elaborazione (16)
Software Security (8)
Laboratorio di Programmazione (6)
  • MSc Thesis - Advisor, "Comparing Quality of Human-written and AI-generated Code", Andrea Scognamiglio, M63001431, 2024/25
  • MSc Thesis - Advisor, "Analisi dell’impatto del training e delle caratteristiche del modello sui generatori di codice basati su IA", Fabrizio Palumbo, M63001665, 2024/25
  • MSc Thesis - Advisor, "Artificial Firmware Designer Assistant: Valutazione del Codice VHDL Generato da AI", Marco Di Fiandra, M63001444, 2023/2024
  • MSc Thesis - Advisor, "Un Metodo per l’Analisi di Bug Residui in Sistemi Python Complessi", Lavinia Russo, M63001479, 2023/2024
  • MSc Thesis - Advisor, "Generazione di Codice VHDL da Linguaggio Naturale: Fine-Tuning di Modelli AI Open-Source e Confronto con Modelli Closed-Source", S.Ten. Carlo Portosalvo, M63001538, 2023/2024
  • MSc Thesis - Advisor, "Valutazione delle Tecnologie di Generazione Automatica di Codice per Firmware: Performance dei Modelli AI Closed-Source", Anna Scala, M63000662, 2023/2024
  • MSc Thesis - Advisor, "Security Testing of AI Code Generators", Alessio Foggia, M63001311, 2022/2023
  • MSc Thesis - Advisor, "Software Vulnerability Analysis for AI-generated code", Roberta De Luca, M63001270, 2022/2023
  • MSc Thesis - Co-Advisor, "Context-Aware Generation of Software Exploits via AI-based Code Generators", Martina Russo, M63001128, 2021/22
  • MSc Thesis - Co-Advisor, "Towards the Automatic Evaluation of Code Syntax and Semantics", Emiliano Fiorenza, M63000920, 2021/22
  • MSc Thesis - Co-Advisor, "Leveraging Adversarial Training to improve NMT Models Robustness in Code Generation tasks", Cristina Improta, M63001057, 2020/21
  • MSc Thesis - Co-Advisor, "Rilevamento di Anomalie in Sistemi Cloud: un Approccio Automatizzato Basato su Distributed Tracing", Nicola Apa, M63001050, 2020/21
  • MSc Thesis - Co-Advisor, "Analisi Empirica per il Rilevamento delle Anomalie su OpenStack in Scenari Multi-Tenant", Roberto Scarpati, M63001013, 2020/21
  • MSc Thesis - Co-Advisor, "Leveraging Neural Machine Translation to Automatically Generate Software Exploits", Simona De Vivo, M63001022, 2020/21. The thesis was evaluated as one of the three best master's degree theses of the degree courses in Computer Science, Computer Engineering or Data Science of the universities of Campania, on research topics related to Big Data Analytics, Machine Learning and Multimedia in the periodo between 01/07 /2020 and 30/12/2021 (award in memory of “Prof. Antonio Picariello").
  • MSc Thesis - Co-Advisor, "Enhancing Failure Analysis of Cloud Infrastructures by using Deep Learning", Gabriella Karamanolis, M63000754, 2019/20
  • MSc Thesis - Co-Advisor, "Runtime Verification via Stream Processing in Cloud Computing Infrastructures", Angela Scibelli, M63000706, 2019/20

Publications

A selection of journal articles and conference papers that reflect the main directions of my ongoing research.

For the full list of publications, please visit my Google Scholar profile.

  • Feng, R., Pearce, H., Liguori, P., and Sui, Y. (2025). CGP-Tuning: Structure-Aware Soft Prompt Tuning for Code Vulnerability Detection. IEEE Transactions on Software Engineering.
    DOI
  • Improta, I., Liguori, P., Natella, R., Cukic, B., and Cotroneo, D. (2025). Enhancing robustness of AI offensive code generators via data augmentation. Empirical Software Engineering.
    DOI
  • Cotroneo, D., De Luca, R., and Liguori, P. (2025). DeVAIC: A tool for security assessment of AI-generated code. Information and Software Technology.
    DOI
  • Cotroneo, D., Improta, C., and Liguori, P. (2025). Human-Written vs. AI-Generated Code: A Large-Scale Study of Defects, Vulnerabilities, and Complexity. IEEE 36th International Symposium on Software Reliability Engineering (ISSRE).
    DOI
  • Improta, C., Tufano, R., Liguori, P., Cotroneo, D., and Bavota, G. (2025). Quality In, Quality Out: Investigating Training Data's Role in AI Code Generation. 33rd IEEE/ACM International Conference on Program Comprehension (ICPC).
    DOI
  • Cotroneo, D., Improta, C., Liguori, P., and Natella, R. (2024) Vulnerabilities in AI Code Generators: Exploring Targeted Data Poisoning Attacks. 32nd IEEE/ACM International Conference on Program Comprehension (ICPC).
    DOI
  • Cotroneo, D., De Rosa, G., and Liguori, P. (2025). PyResBugs: A Dataset of Residual Python Bugs for Natural Language-Driven Fault Injection. In 2025 IEEE/ACM Second International Conference on AI Foundation Models and Software Engineering (Forge).
    DOI

Collaborations & Service

International partnerships, editorial roles, and program leadership

Academic Collaborations

  • University of North Carolina at Charlotte
    AI offensive code generation from natural language descriptions. Collaboration with Prof. Domenico Cotroneo and Prof. Bojan Cukic
  • USI Lugano
    Quality assessment of AI-generated code. Collaboration with Prof. Gabriele Bavota
  • University of Coimbra (Portugal)
    Vulnerability injection in AI-generated code using LLMs. Collaboration with Prof. Naghmeh Ramezani Ivaki

Program Leadership

  • Program Chair
    ReSAISE 2023–2025 (Reliable and Secure AI for Software Engineering), co-located with ISSRE

Editorial Service

  • Guest Editor
    Journal of Systems and Software
    Special Issue: Reliable and Secure Large Language Models for Software Engineering
  • Guest Editor
    Automated Software Engineering
    Special Issue: Reliable and Secure AI-based Code Generators

Community Engagement

  • IFIP WG 10.4
    Dependable Computing — invited member (2024–2025 meetings in Australia, USA, Brazil)
  • IEEE/ACM PC Member
    ISSRE, DSN, FORGE, EDCC, NLBSE, LADC
  • Reviewer for
    TSE, TDSC, EMSE, IST, TOSEM, JSS

Contact Information

Get in touch for research collaboration or academic inquiries

Pietro Liguori

Assistant Professor

Department of Electrical Engineering and Information Technology (DIETI)

University of Naples Federico II

pietro.liguori@unina.it

Via Claudio, 21, 80125 Napoli NA, Italy