From Code Generation Towards Software Engineering: Advancing Code Intelligence w/ Language Models

Large language models (LLMs) have revolutionized the way how humans write code, but they still have limitations in comprehensively reasoning about software programs to assist with more involved software engineering tasks. In this talk, I will first provide an overview of my research on enhancing LLMs' code intelligence, optimizing each step of the development pipeline towards more complex software engineering tasks. I will then delve into my key contributions, focusing on how to equip LLMs with (1) symbolic reasoning for program semantics and (2) global reasoning for software dependencies. Finally, I will discuss the future of AI-driven software engineering, envisioning a path to approach full-stack automation in a trustworthy way.