📄 中文摘要
不良媒体筛查是金融机构反洗钱（AML）和客户尽职调查（KYC）合规流程中的关键环节。传统方法依赖于基于关键词的搜索，导致高假阳性率或需要大量人工审核。提出了一种代理系统，该系统利用大型语言模型（LLMs）和检索增强生成（RAG）技术来自动化不良媒体筛查。该系统实施了多步骤的方法，其中LLM代理在网络上搜索，检索和处理相关文档，并为每个对象计算不良媒体指数（AMI）分数。通过在包含政治曝光人士（PEPs）数据集上使用多个LLM后端对该方法进行了评估。
用于反洗钱合规的代理型大型语言模型框架在不良媒体筛查中的应用
📄 English Summary
An Agentic LLM Framework for Adverse Media Screening in AML Compliance
Adverse media screening is a critical component of anti-money laundering (AML) and know-your-customer (KYC) compliance processes in financial institutions. Traditional approaches rely on keyword-based searches that generate high false-positive rates or require extensive manual review. An agentic system is presented that leverages Large Language Models (LLMs) with Retrieval-Augmented Generation (RAG) to automate adverse media screening. The system implements a multi-step approach where an LLM agent searches the web, retrieves and processes relevant documents, and computes an Adverse Media Index (AMI) score for each subject. The approach is evaluated using multiple LLM backends on a dataset comprising Politically Exposed Persons (PEPs).