An Empirical Evaluation of Using Large Language Models to Recommend Exception Handling Tasks accepted early at SBES’26.