Large language model (LLM) agents are increasingly employing retrieval-augmented generation (RAG) to improve the factuality of their responses. However, in practice, these systems often need to handle...

Unit tests (UTs) play an instrumental role in assessing code correctness as well as providing feedback to a large language model (LLM) as it iteratively debugs faulty code, motivating automated test g...