Improving the performance of large language models (LLMs) in complex question-answering (QA) scenarios has always been a research focal point. Recent studies have attempted to enhance LLMs' performance by combining step-wise planning with external retrieval. While effective for advanced models like GPT-3.5, smaller LLMs face challenges in decomposing complex questions, necessitating supervised fine-tuning. Previous work has relied on manual annotation and knowledge distillation from teacher LLMs, which are time-consuming and not accurate enough. In this paper, we introduce a novel framework for enhancing LLMs' planning capabilities by using planning data derived from knowledge graphs (KGs). LLMs fine-tuned with this data have improved planning capabilities, better equipping them to handle complex QA tasks that involve retrieval. Evaluations on multiple datasets, including our newly proposed benchmark, highlight the effectiveness of our framework and the benefits of KG-derived planning data.

Matthew B. Matheny, 40, of Warren, Ohio, appeared to be in good health when he was found on Mount St. Helens, authorities say. An Ohio man who went missing last week after setting out on a hike on Mount St. Helens in Washington state, was found alive Wednesday -- having survived by eating bees and berries, authorities and the man's relatives said. Matthew B. Matheny, 40, of Warren, Ohio, was reported missing by friends after he failed to return from his hike along Blue Lake Trail on the southwest side of St. Helens. His parents told reporters that he was not familiar with the terrain and got lost. They said he had not seen anyone since Aug. 9. Dozens of search-and-rescue personnel, assisted by helicopters, tracking dogs and a drone operated by the local sheriff's office searched daily for Matheny.