Researchers have developed a novel “Retrieve-Rewrite-Answer” framework to improve the performance of large language models (LLMs) in Knowledge Graph Question Answering (KGQA). The three-stage approach first fetches pertinent Knowledge Graph (KG) data, then converts it into well-textualized statements, which are finally used for answering complex questions. Unique features include an “answer-sensitive” KG-to-Text methodology and an automatic corpus generation method using ChatGPT, addressing issues like data scarcity. Rigorous testing against various benchmarks and existing LLMs revealed that the framework consistently outperforms existing methods, particularly excelling with the T5 model.
For further actions, you may consider blocking this person and/or reporting abuse