The first CREATE SE4AI Trainee Talks of the fall 2024 semester was held on October 16. During this session, Master's candidate Lin Ling presented her research titled "Bias Unveiled: Investigating Social Bias in LLM-Generated Code". Following her presentation, Doctoral candidate Junjie Li shared his study on "An Exploratory Study on Fine-Tuning Large Language Models for Secure Code Generation". Both trainees are under the supervision of Prof. Jinqiu Yang at Concordia University.

The webinar recording is available to view on our CREATE SE4AI YouTube channel.

On Tuesday, March 25th, at 11:00 AM EDT, we welcomed Queen’s University Ph.D. student Adekunle Ajibode. Adekunle shared his research on “Towards Semantic Versioning of Open Pre-trained Language Model Releases on Hugging Face,” which was recently published in the EMSE journal: https://arxiv.org/pdf/2409.10472.

Immediately after, Aurel Ikama Honey, a Ph.D. student at Polytechnique Montréal, presented the preliminary results of his work on “A Comparative Study of LLM Architectures for Root Cause Analysis in End-to-End Test Logs.”

Each trainee spoke for 20 minutes, followed by a 10-minute Q&A session.

The webinar recording is available to view on our CREATE SE4AI YouTube channel.