arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2502.15865q-fin.GNcs.AIcs.CL
收藏

标准基准测试失效——审计金融领域的LLM智能体必须优先考虑风险

Standard Benchmarks Fail -- Auditing LLM Agents in Finance Must Prioritize Risk

University of California, Santa Barbara(加州大学圣巴巴拉分校);Georgia Institute of Technology(佐治亚理工学院);Nanyang Technological University(南洋理工大学)

Zichen Chen, Jiaao Chen, Jianda Chen, Misha Sra

AI总结:

本文指出金融LLM智能体的标准基准测试只重表现而忽视安全风险,提出以风险为核心的三级压力测试框架,并通过审计六个智能体揭示隐藏漏洞,呼吁将安全预算作为首要评估标准。

摘要:

标准基准测试专注于衡量大语言模型(LLM)智能体在金融领域的表现有多好,却几乎不涉及它们是否可以安全部署。我们认为,准确率指标和基于收益的评分提供了一种可靠性的假象,忽视了诸如幻觉事实、过时数据以及对抗性提示操纵等脆弱性。我们持明确立场:金融LLM智能体的评估首先且最重要的是基于其风险状况,而非其点估计性能。借鉴风险工程原理,我们提出了一个三级议程:模型、工作流和系统,用于在现实故障模式下对LLM智能体进行压力测试。为说明这一转变为何迫在眉睫,我们审计了六个基于API和开放权重的LLM智能体在三项高影响力任务上的表现,并揭示了传统基准测试所忽略的隐藏弱点。最后,我们为研究人员、从业者和监管机构提出了可操作的建议:在未来的研究中审计风险感知指标,在发布数据集的同时发布压力场景,并将“安全预算”视为首要成功标准。只有重新定义什么是“好”,社区才能负责任地推进AI驱动的金融发展。

英文摘要:

Standard benchmarks fixate on how well large language model (LLM) agents perform in finance, yet say little about whether they are safe to deploy. We argue that accuracy metrics and return-based scores provide an illusion of reliability, overlooking vulnerabilities such as hallucinated facts, stale data, and adversarial prompt manipulation. We take a firm position: financial LLM agents should be evaluated first and foremost on their risk profile, not on their point-estimate performance. Drawing on risk-engineering principles, we outline a three-level agenda: model, workflow, and system, for stress-testing LLM agents under realistic failure modes. To illustrate why this shift is urgent, we audit six API-based and open-weights LLM agents on three high-impact tasks and uncover hidden weaknesses that conventional benchmarks miss. We conclude with actionable recommendations for researchers, practitioners, and regulators: audit risk-aware metrics in future studies, publish stress scenarios alongside datasets, and treat ``safety budget'' as a primary success criterion. Only by redefining what ``good'' looks like can the community responsibly advance AI-driven finance.

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。

↑