arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2309.08922cs.CL
收藏

通过工具对话与高效微调大语言模型实现多模态多跳问答

Multimodal Multi-Hop Question Answering Through a Conversation Between Tools and Efficiently Finetuned Large Language Models

University of Waterloo(滑铁卢大学);Mila(米拉人工智能研究院);Université de Montréal(蒙特利尔大学)

Hossein Rajabzadeh, Suyuchen Wang, Hyock Ju Kwon, Bang Liu

AI总结:

本文提出一种工具交互的分而治之策略,将多模态多跳问题分解为单模态单跳子问题交由工具回答,并利用chatGPT生成数据集高效微调大语言模型,在两个复杂问答数据集上显著超越现有方法。

摘要:

我们采用一种工具交互的分而治之策略,使大语言模型(LLMs)能够回答复杂的多模态多跳问题。具体而言,我们利用大语言模型的能力,将给定的多模态多跳问题分解为单模态单跳子问题,并由预定义工具集中的适当工具来回答。在所有相应工具向大语言模型提供答案后,大语言模型生成下一个相关的单模态单跳子问题。为了增强大语言模型的推理能力,我们提示chatGPT生成一个工具交互的分而治之数据集。该数据集随后被用于高效微调相应的大语言模型。为评估该方法的有效性,我们在两个近期提出的复杂问答数据集上进行了评测。实验分析表明,与现有最先进解决方案相比取得了显著改进,表明了我们策略的有效性和通用性。

英文摘要:

We employ a tool-interacting divide-and-conquer strategy enabling large language models (LLMs) to answer complex multimodal multi-hop questions. In particular, we harness the power of large language models to divide a given multimodal multi-hop question into unimodal single-hop sub-questions to be answered by the appropriate tool from a predefined set of tools. After all corresponding tools provide the LLM with their answers, the LLM generates the next relevant unimodal single-hop question. To increase the reasoning ability of LLMs, we prompt chatGPT to generate a tool-interacting divide-and-conquer dataset. This dataset is then used to efficiently finetune the corresponding LLM. To assess the effectiveness of this approach, we conduct an evaluation on two recently introduced complex question-answering datasets. The experimental analysis demonstrate substantial improvements over existing state-of-the-art solutions, indicating the efficacy and generality of our strategy

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑