DEEPSE Forum Seminars - Assessing AI Assistants in Code Generation Tasks
Speaker: Prof. Leonardo Mariani
DEIB - Beta Room (Bld. 24)
On line by Teams
June 20th, 2025 | 2.30 pm
Contact: Prof. Giovanni Quattrocchi
DEIB - Beta Room (Bld. 24)
On line by Teams
June 20th, 2025 | 2.30 pm
Contact: Prof. Giovanni Quattrocchi
Abstract
On June 20th, 2025 at 2.30 pm a new appointment of DEEPSE Forum Seminars series titled "Assessing AI Assistants in Code Generation Tasks" will take place at DEIB Beta Room (Building 24) and on line by Teams.
The increasing adoption of AI assistants capable of generating code—such as GitHub Copilot, ChatGPT, and similar tools—has introduced new opportunities and challenges in software development. This talk presents the results of a series of experiments aimed at evaluating the ability of AI assistants to generate the implementation of Java methods. The talk discusses findings about the correctness and quality of the generated code. It also reports results about the impact of prompts and configurations on the resulting code. Finally, it discusses risks concerning the possible generation and unintended reuse of code protected by copyleft licenses.
The increasing adoption of AI assistants capable of generating code—such as GitHub Copilot, ChatGPT, and similar tools—has introduced new opportunities and challenges in software development. This talk presents the results of a series of experiments aimed at evaluating the ability of AI assistants to generate the implementation of Java methods. The talk discusses findings about the correctness and quality of the generated code. It also reports results about the impact of prompts and configurations on the resulting code. Finally, it discusses risks concerning the possible generation and unintended reuse of code protected by copyleft licenses.
Short Bio
Leonardo Mariani is Full Professor at the University of Milano - Bicocca. He holds a Ph.D. in Computer Science received from the same university in 2005. His research interests include software engineering, in particular software testing, program analysis, automated debugging, AI4SE, and self-adaptive systems. He has authored more than 150 papers appeared at top software engineering conferences and journals. In 2023, he received the ICST MIP award for his ICST 2013 paper. He has been awarded with the ERC Consolidator Grant in 2015, an ERC Proof of Concept grant in 2018, and he is currently active in several national and international projects. He is regularly involved in the organization of major software engineering conferences.
