Les séminaires du GEODES visent à rassembler les membres du groupe de manière périodique dans un cadre semi-formel pour échanger autour d'une thématique.
Les séminaires donnent aux étudiants l'opportunité de partager leur avancement et trouvaille et présenter un sujet lié aux thèmes de recherche du groupe. Ils ont ainsi l'occasion de pratiquer leur présentation à venir et récolter de la rétroaction.
Discussions à venir
-
Agentic AI for Software: Lessons in Trust from AutoCodeRover- Abhik Roychoudhury – Professor at National Univ. of Singapore
Abstract:
AI agents have recently shown significant promise in software engineering. Much public attention has been transfixed on the topic of code generation from Large Language Models (LLMs) via a prompt. However, software engineering is much more than programming, and AI agents go far beyond instructions given by a prompt. Conceptually, the key to successfully developing trustworthy agentic AI-based software workflows will be to resolve the core difficulty in software engineering - the deciphering and clarification of developer intent. Specification inference, or deciphering the intent, thus lies at the heart of many software tasks, including software maintenance and program repair. A successful deployment of agentic technology into software engineering would involve making conceptual progress in such intent inference via agents. We discuss, to some length, the AutoCodeRover agent which embodies such intent inference. The agent has been integrated into the SonarQube static analysis tool, which is used by many enterprise customers. Trusting the AI agent becomes a key aspect, for coding agents to be put into production. Higher automation also leads to higher volume of code being automatically generated, and then integrated into code-bases. To deal with this explosion, an emerging direction is AI-based verification and validation (V & V) of AI generated code. We posit that agentic software workflows in future will include such AI-based V&V.
Bio:
Abhik Roychoudhury is Provost's Chair Professor of Computer Science at the National University of Singapore (NUS), where he leads a research team on Trustworthy and Secure Software (TSS). He is Senior Advisor at SonarSource, subsequent to the acquisition of his spinoff AutoCodeRover on AI agents for coding. He received his PhD in Computer Science from the Stony Brook University in 2000, and has been a faculty member at NUS School of Computing since 2001. Abhik's group at NUS has focused on symbolic program analysis, along with applications of program analysis to areas such as computer security, agentic AI or cyber-physical systems. These works have been honored with various awards including an International Conference on Software Engineering (ICSE) Most Influential Paper Award (Test-of-time award) for symbolic analysis based program repair, IEEE New Directions Award 2022 (jointly with Cristian Cadar) for contributions to symbolic execution. Abhik was the inaugural recipient of the NUS Outstanding Graduate Mentor Award. Doctoral students graduated from his research team have taken up faculty positions in many top academic institutions, and they have gone on to receive many prestigious early career awards (including ACM-W Rising Star Award given to only one female faculty member in Computing). Abhik has served the software engineering research community in various capacities including as chair of the major conferences (ICSE and FSE), as well as chair of the FSE steering committee. He is the current Editor-in-Chief of the ACM Transactions on Software Engineering and Methodology (TOSEM), and a member of the editorial board of Communications of the ACM. Abhik is a Fellow of the ACM, recognized for contributions to automated program repair and fuzz testing.Thursday, November 13, 2025 @15h00 3195, Pav. André-Aisenstadt
Dernières discussions
-
Making Models Great Again: Model Construction in the Age of Large Language Models
- Mehrdad Sabetzadeh – Professor at University of Ottawa
This seminar explores how large language models are revitalizing model-driven engineering, examining their potential to make modeling more efficient, practical, and competitive with direct code development.
2025-10-29@14h00 6214, Pav. André-Aisenstadt
Discussions antérieures
-
Multilinguality in Large Language Models [presentation]
- Ayla Rigouts Terryn – Assistant Professor at Dép. de Linguistique et traduction, UdeM
-
Verification of Safety-Critical Systems
- Daniel Varro – Professor of Software Engineering at Linkoping University
-
Recent Progress in Neural Program Repair
- Martin Monperrus – Professor of Software Technology at KTH Royal Institute of Technology
-
Extending Scala for Safe Concurrent Programming
- Philipp Haller – Associate professor of computer science at KTH Royal Institute of Technology
-
The IDEA of Us: Engineering Socially-Aware AI Systems
- Amel Bennaceur – Associate Professor at the Open University, UK
-
Embracing Deep Variability For Reproducibility & Replicability
- Benoit Combemale – Full Professor of Software Engineering at the University of Rennes
-
AI-Assisted Automated Programing and Automated Program Repair
- He Ye – Postdoctoral researcher
-
Large Language Models for Code - Introduction and Recent Breakthroughs
- Martin WEYSSOW – PhD candidate
-
Open collaborative data engineering
- Prof. Dirk Riehle – Friedrich-Alexander-Universität Erlangen-Nürnberg
-
Video Game Software Engineering: Troubles, Techniques, and Tools
- Cristiano POLITOWSKI – Postdoc
-
Conceptualization and Development of ML-based Recommender Systems for Software Engineering [presentation]
- Claudio DI SIPIO – Invited researcher
-
Characterizing and building datasets of software repositories
- Jessie GALASSO – Postdoc
-
A Multi-Step Learning Approach to Assist Code Review
- Oussama BEN SGHAIER – PhD student
-
Migrating Enterprise applications into Software Product Lines
- Tewfik ZIADI – Paris-Sorbonne University, France
- Uncertainty in Software Models [presentation]
- Learning from Code Repositories to Recommend Model Classes
- Process modeling, analysis and enactment with the FTG+PM formalism [presentation]
- Systematic Reviews with ReLiS [presentation]
- Fine-grained Analysis of Similar Code Snippets
- The ongoing problem of reuse and evolution in software architectures: illustration through four research projects
-
Moose – A modular, composable software analysis environment
- Stéphane Ducasse – INRIA Lille, France
- Nicolas Anquetil – University of Lille, France
- Introduction to Reinforcement Learning
- Multi-paradigm modeling of complex heterogeneous systems – Are we there yet?
-
Improving online visibility for you and your research [presentation]
- Bentley James Oakes
- Lola Burgueño – Open University of Catalonia, Spain
-
AST-Probe: Recovering abstract syntax trees from hidden representations of pre-trained language models
- José Antonio Hernández López – Universidad de Murcia, Spain
Contact
Got an interesting topic you'd like to present? Reach out to your favorite postdoc for details.
