{"ID":2874084,"CreatedAt":"2026-06-01T04:54:23.091178241Z","UpdatedAt":"2026-06-01T04:54:23.091178241Z","DeletedAt":null,"paper_url":"https://arxiv.org/abs/2509.04770","arxiv_id":"2509.04770","title":"Research on Multi-hop Inference Optimization of LLM Based on MQUAKE Framework","abstract":"Accurately answering complex questions has consistently been a significant challenge for Large Language Models (LLMs). To address this, this paper proposes a multi-hop question decomposition method for complex questions, building upon research within the MQUAKE framework. Utilizing the LLAMA3 model, we systematically investigate the impact of multi-hop question decomposition within knowledge graphs on model comprehension and reasoning accuracy, both before and after model training. In our experiments, we systematically partitioned and converted the MQUAKE-T dataset into two distinct formats: a single-hop dataset designed for directly answering complex questions, and a multi-hop dataset constructed using the multi-hop question decomposition method. We then fine-tuned the LLAMA3 model on these datasets and conducted inference tests. Our results demonstrate that, without fine-tuning the LLM, the prediction performance based on the multi-hop question decomposition method significantly outperforms the method of directly answering complex questions. After fine-tuning using the LoRA (Low-Rank Adaptation) method, the performance of both approaches improved compared to the untrained baseline. Crucially, the method utilizing multi-hop decomposition consistently maintained its superiority. These findings validate the effectiveness of the multi-hop decomposition method both before and after training, demonstrating its capability to effectively enhance the LLM's ability to answer complex questions.","short_abstract":"Accurately answering complex questions has consistently been a significant challenge for Large Language Models (LLMs). To address this, this paper proposes a multi-hop question decomposition method for complex questions, building upon research within the MQUAKE framework. Utilizing the LLAMA3 model, we systematically i...","url_abs":"https://arxiv.org/abs/2509.04770","url_pdf":"https://arxiv.org/pdf/2509.04770v1","authors":"[\"Zucheng Liang\",\"Wenxin Wei\",\"Kaijie Zhang\",\"Hongyi Chen\"]","published":"2025-09-05T02:58:45Z","proceeding":"cs.CL","tasks":"[\"cs.CL\",\"cs.LG\"]","methods":"[\"Large Language Model\",\"Language Model\",\"LoRA\"]","has_code":false}