Software Reliability in the Era of Large Language Models: A Dual Perspective
Speaker: David Lo – Singapore, SingaporeTopic(s): Artificial Intelligence, Machine Learning, Computer Vision, Natural language processing , Software Engineering and Programming
Abstract
Much software engineering research has been dedicated to building reliable software systems. The last two decades have witnessed the growth of software engineering data availability that spurred many AI-driven automated solutions. The last few years saw rapid growth in the construction of specialized solutions based on Large Language Models (LLM) to aid software engineers in many tasks, including improving software reliability. However, LLM has unique challenges, presenting new reliability concerns that must be managed. This underscores two compelling and complementary research trajectories: Large Language Models for Software Reliability (LLM4SR) and Software Reliability for Large Language Models (SR4LLM). This lecture will showcase promising LLM4SR solutions, focusing on vulnerability repair and runtime error recovery. It will then discuss some reliability issues that affect LLM and preliminary solutions to manage them, highlighting much research needed in SR4LLM. The lecture will conclude with a discussion on future directions, showing how SR and LLM can significantly change software engineering in the years ahead.About this Lecture
Number of Slides: 90 - 110Duration: 60 - 75 minutes
Languages Available: English
Last Updated:
Request this Lecture
To request this particular lecture, please complete this online form.
Request a Tour
To request a tour with this speaker, please complete this online form.
All requests will be sent to ACM headquarters for review.