Towards Effectively Leveraging Execution Traces for Program Repair with Code LLMs

Mirazul Haque, Petr Babkin, Farima Farmahinifarahani +1

Large Language Models (LLMs) show promising performance on various programming tasks, including Automatic Program Repair (APR). However, most approaches to LLM-based APR are limite…