Large language models (LLMs) have made significant strides in various tasks, yet they often struggle with complex reasoning and exhibit poor performance in scenarios where knowledge traceability, timeliness, and accuracy are crucial. To address these limitations, we present Think-on-Graph (ToG), a novel framework that leverages knowledge graphs to enhance LLMs' ability for deep and responsible reasoning. By employing ToG, we can identify entities relevant to a given question and conduct exploration and reasoning to retrieve related triples from an external knowledge database. This iterative procedure generates multiple reasoning pathways consisting of sequentially connected triplets until sufficient information is gathered to answer the question or the maximum depth is reached. Through experiments on complex multi-hop reasoning question-answering tasks, we demonstrate that ToG outperforms existing methods, effectively addressing the aforementioned limitations of LLMs without incurring additional training costs.

翻译：大语言模型（LLMs）在各种任务中取得了显著进展，但在知识可追溯性、时效性和准确性至关重要的场景中，它们往往难以进行复杂推理，且表现不佳。为解决这些局限性，我们提出了Think-on-Graph（ToG），一种利用知识图谱增强大语言模型深度与负责任推理能力的新型框架。通过采用ToG，我们可以识别与给定问题相关的实体，并从外部知识数据库中探索和推理以检索相关三元组。此迭代过程生成由顺序连接的三元组组成的多条推理路径，直至收集到足够信息来回答问题或达到最大深度。通过在复杂多跳推理问答任务上的实验，我们证明ToG优于现有方法，有效解决了大语言模型上述局限性，且无需额外训练成本。