大模型在投资建议中常忽略用户全貌，只依赖风险偏好做决策。
One Size Fits None: Heuristic Collapse in LLM Investment Advice
- 用可解释代理模型分析大模型输出，发现决策高度依赖风险容忍度。
- 其他重要因素如收入、年龄等对决策影响微弱，仅占10%以下贡献。
- 网络搜索缓解但未消除此问题，需专门审计输入敏感性。
大型语言模型日益被用于高风险领域，如医疗咨询、法律解读和金融产品推荐，其中优质建议需综合用户全部背景信息，而非仅关注表面特征。本文研究前沿大模型是否真能实现这一目标，或是否存在启发式坍塌——即复杂多因素决策被简化为少数主导输入。研究聚焦于投资建议场景，该领域法律要求基于客户全面情况个性化推理。通过应用可解释的代理模型分析大模型输出，发现系统性启发式坍塌：投资配置决策主要由自述风险容忍度决定，而其他相关因素贡献极小（平均贡献率低于10%）。进一步发现，网络搜索虽部分缓解此现象，但未能根本解决。结果表明，仅靠模型规模提升或网络搜索增强无法消除启发式坍塌，部署大模型作为顾问时，必须审计其对输入的敏感性，而不仅关注输出质量。
原文摘要 · Abstract (English)
Large language models are increasingly deployed as advisors in high-stakes domains -- answering medical questions, interpreting legal documents, recommending financial products -- where good advice requires integrating a user's full context rather than responding to salient surface features. We investigate whether frontier LLMs actually do this, or whether they instead exhibit heuristic collapse: a systematic reduction of complex, multi-factor decisions to a small number of dominant inputs. We study the phenomenon in investment advice, where legal standards explicitly require individualized reasoning over a client's full circumstances. Applying interpretable surrogate models to LLM outputs, we find systematic heuristic collapse: investment allocation decisions are largely determined by self-reported risk tolerance, while other relevant factors contribute minimally. We further find that web search partially attenuates heuristic collapse but does not resolve it. These findings suggest that heuristic collapse is not resolved by web search augmentation or model scale alone, and that deploying LLMs as advisors requires auditing input sensitivity, not just output quality.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。