通过自选择优化检索增强生成中的知识整合
Optimizing Knowledge Integration in Retrieval-Augmented Generation with Self-Selection
AI总结：
针对检索增强生成（RAG）中外部检索知识与LLM内部参数化知识整合的挑战，提出自选择RAG框架及Self-Selection-RGP方法，通过DPO训练提升LLM的生成与选择能力，在NQ和TrivialQA数据集上优于基线方法。
摘要：
检索增强生成（RAG）将外部知识整合到大型语言模型（LLMs）中，已被证明能有效提升LLMs生成更准确可靠的响应。然而，如何有效将外部检索到的知识与LLMs内部的参数化知识整合仍是一项重大挑战。在本研究中，我们提出一种新颖的自选择RAG框架，让LLM从仅使用内部参数化知识生成的响应和结合外部检索知识生成的响应这两个配对响应中进行选择，以提高准确性。为此，我们设计了Self-Selection-RGP方法，通过在精心构建的检索生成偏好（RGP）数据集上使用直接偏好优化（DPO）训练LLM，增强其生成和选择正确答案的能力。在Natural Questions（NQ）和TrivialQA数据集上使用两个开源LLM（即Llama2-13B-Chat和Mistral-7B）的实验结果充分证明了我们的方法优于其他基线方法。
英文摘要：
Retrieval-Augmented Generation (RAG), which integrates external knowledge into Large Language Models (LLMs), has proven effective in enabling LLMs to produce more accurate and reliable responses. However, it remains a significant challenge how to effectively integrate external retrieved knowledge with internal parametric knowledge in LLMs. In this work, we propose a novel Self-Selection RAG framework, where the LLM is made to select from pairwise responses generated with internal parametric knowledge solely and with external retrieved knowledge together to achieve enhanced accuracy. To this end, we devise a Self-Selection-RGP method to enhance the capabilities of the LLM in both generating and selecting the correct answer, by training the LLM with Direct Preference Optimization (DPO) over a curated Retrieval Generation Preference (RGP) dataset. Experimental results with two open-source LLMs (i.e., Llama2-13B-Chat and Mistral-7B) well demonstrate the superiority of our approach over other baseline methods on Natural Questions (NQ) and TrivialQA datasets.
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。
