arXivDaily arXiv每日学术速递 周一至周五更新
​通知:联邦节假日调休:哥伦布日 / 原住民日(Columbus Day / Indigenous Peoples' Day),今日(13日)无更新。
arXiv 2507.16711cs.CLcs.AI

推进风险与质量保证:用于改进法规合规性的 RAG 聊天机器人

Advancing Risk and Quality Assurance: A RAG Chatbot for Improved Regulatory Compliance

Fraunhofer IAIS(弗劳恩霍夫智能分析和信息系统研究所);University of Bonn(波恩大学);PricewaterhouseCoopers GmbH(普华永道有限责任公司);Lamarr Institute(拉马尔研究所)

Lars Hillebrand, Armin Berger, Daniel Uedelhoven, David Berghaus, Ulrich Warning, Tim Dilmaghani, Bernd Kliem, Thomas Schmid, Rüdiger Loitz, Rafet Sifa

AI总结:

本文提出一种结合大语言模型、混合搜索与相关性增强的 RAG 系统,用于改进高监管行业风险与质量查询处理,并在 124 条专家标注查询上验证其优于传统 RAG 方法。

摘要:

在高度受监管的行业中,风险与质量(R&Q)保证需要不断应对复杂的监管框架,员工每天处理大量需要准确政策解读的查询。依赖特定专家的传统方法会造成运营瓶颈并限制可扩展性。我们提出了一种新颖的检索增强生成(RAG)系统,利用大语言模型(LLMs)、混合搜索和相关性增强来改进 R&Q 查询处理。该系统已在 124 条专家标注的真实查询上进行了评估,并已实际部署,结果表明其相比传统 RAG 方法有显著改进。此外,我们进行了广泛的超参数分析,以比较和评估多种配置设置,为从业者提供了有价值的见解。

英文摘要:

Risk and Quality (R&Q) assurance in highly regulated industries requires constant navigation of complex regulatory frameworks, with employees handling numerous daily queries demanding accurate policy interpretation. Traditional methods relying on specialized experts create operational bottlenecks and limit scalability. We present a novel Retrieval Augmented Generation (RAG) system leveraging Large Language Models (LLMs), hybrid search and relevance boosting to enhance R&Q query processing. Evaluated on 124 expert-annotated real-world queries, our actively deployed system demonstrates substantial improvements over traditional RAG approaches. Additionally, we perform an extensive hyperparameter analysis to compare and evaluate multiple configuration setups, delivering valuable insights to practitioners.

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。

↑