arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2407.21055cs.CL
收藏

Bailicai:面向医疗应用的领域优化检索增强生成框架

Bailicai: A Domain-Optimized Retrieval-Augmented Generation Framework for Medical Applications

University of South China(南华大学)

Cui Long, Yongbin Liu, Chunping Ouyang, Ying Yu

AI总结:

本文提出面向医疗领域的Bailicai框架,通过四个子模块将检索增强生成与优化的大语言模型集成,在多个医疗基准上超越现有医疗LLMs和GPT-3.5,并有效缓解幻觉与文档噪声问题。

摘要:

大型语言模型(LLMs)在自然语言理解方面展现出卓越的能力,促使人们广泛探索其在各个领域的潜在应用。在医疗领域,开源LLMs经过领域特定微调后表现出中等效果;然而,它们仍显著逊色于GPT-4和GPT-3.5等专有模型。这些开源模型在领域特定知识的全面性方面存在局限,并在文本生成过程中表现出产生“幻觉”的倾向。为缓解这些问题,研究人员采用了检索增强生成(RAG)方法,该方法在保持模型内部参数不变的同时,利用外部知识库中的背景信息增强LLMs。然而,文档噪声可能对性能产生不利影响,且RAG在医疗领域的应用仍处于起步阶段。本研究提出了Bailicai框架:一种将检索增强生成与针对医疗领域优化的大型语言模型相结合的新型集成方案。Bailicai框架通过实现四个子模块来增强LLMs在医学中的性能。实验结果表明,Bailicai方法在多个医疗基准上超越了现有的医疗领域LLMs,并超过了GPT-3.5的性能。此外,Bailicai方法有效减轻了LLMs在医疗应用中普遍存在的幻觉问题,并改善了传统RAG技术在处理不相关或伪相关文档时面临的噪声相关挑战。

英文摘要:

Large Language Models (LLMs) have exhibited remarkable proficiency in natural language understanding, prompting extensive exploration of their potential applications across diverse domains. In the medical domain, open-source LLMs have demonstrated moderate efficacy following domain-specific fine-tuning; however, they remain substantially inferior to proprietary models such as GPT-4 and GPT-3.5. These open-source models encounter limitations in the comprehensiveness of domain-specific knowledge and exhibit a propensity for 'hallucinations' during text generation. To mitigate these issues, researchers have implemented the Retrieval-Augmented Generation (RAG) approach, which augments LLMs with background information from external knowledge bases while preserving the model's internal parameters. However, document noise can adversely affect performance, and the application of RAG in the medical field remains in its nascent stages. This study presents the Bailicai framework: a novel integration of retrieval-augmented generation with large language models optimized for the medical domain. The Bailicai framework augments the performance of LLMs in medicine through the implementation of four sub-modules. Experimental results demonstrate that the Bailicai approach surpasses existing medical domain LLMs across multiple medical benchmarks and exceeds the performance of GPT-3.5. Furthermore, the Bailicai method effectively attenuates the prevalent issue of hallucinations in medical applications of LLMs and ameliorates the noise-related challenges associated with traditional RAG techniques when processing irrelevant or pseudo-relevant documents.

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。

↑