Research Highlights
Estimating Plausibility of LLM-Generated Debugging Feedback
A new paper by Brendan Toscano investigates whether the plausibility of AI-generated debugging feedback for beginner programmers can be predicted using computable features, achieving up to 92% accuracy in flagging helpful feedback.
See more on the CILS Instagram page
DAHLIA Intelligence Inc.
Supervisor: Dr. Andrew McIntyre
Interns: Luke Wagner & Luke Mainwaring
This Mitacs Accelerate Entrepreneur project investigates autonomous AI agents that interpret natural language requests, create multi-step plans, and execute actions across email, calendar, and reminder systems.
