Researchers have developed a framework to help healthcare chatbots better understand patient queries by asking targeted follow-up questions to clarify missing context. The knowledge-guided agentic framework interprets the initial query and uses a task-specific knowledge graph to identify plausible hypotheses, then asks follow-up questions to gather necessary information before providing an answer. This approach was evaluated on two benchmarks and showed significant improvements in accuracy compared to direct prompting or rephrasing queries without additional context. --- Why it matters: This matters because healthcare chatbots often struggle with ambiguous patient queries, leading to inaccurate or incomplete responses. By clarifying the context through targeted follow-up questions, this framework can improve the reliability of these systems and provide better care for patients. Source: https://arxiv.org/abs/2608.19875