arXiv 2411.138002025-10-07cs.CL

解释GPT的抑郁症图式：一种机器行为分析

Explaining GPTs' Schema of Depression: A Machine Behavior Analysis

Stony Brook University（石溪大学）；Carnegie Mellon University（卡内基梅隆大学）；Lund University（吕勒欧大学）；Stanford University（斯坦福大学）；Institute of Human-Centered AI at Stanford University（斯坦福大学人机交互人工智能研究所）；The University of Texas at Dallas（德克萨斯大学达拉斯分校）；Vanderbilt University（范德比大学）；University of Bonn（波恩大学）；The Lamarr Institute for Machine Learning & Artificial Intelligence（拉马尔人工智能与机器学习研究所）

Adithya V Ganesan, Vasudha Varadarajan, Yash Kumar Lal, Veerle C. Eijsbroek, Katarina Kjell, Oscar N. E. Kjell, Tanuja Dhanasekaran, Elizabeth C. Stade, Johanne… 展开作者

Adithya V Ganesan, Vasudha Varadarajan, Yash Kumar Lal, Veerle C. Eijsbroek, Katarina Kjell, Oscar N. E. Kjell, Tanuja Dhanasekaran, Elizabeth C. Stade, Johannes C. Eichstaedt, Ryan L. Boyd, H. Andrew Schwartz, Lucie Flek

备注：: 25 pages, 1 table, 4 figures, 1 supplementary table, 5 supplementary figures, 59 references

阅读 PDF

arXiv 页面 ↗HTML 原文 ↗

AI总结：

本研究通过机器行为分析，利用测量理论解码GPT-4和GPT-5对抑郁症症状的关联图式，发现其具有收敛效度但存在症状权重偏差，并提出可推广的模型可解释性方法。

摘要：

使用诸如ChatGPT（GPT-4/GPT-5）等大型语言模型进行心理健康支持的应用增长迅速，成为评估和帮助患有抑郁症等情绪障碍人群的一条有前景的途径。然而，我们对这些语言模型关于精神障碍的图式（即它们如何在内部关联和解释此类障碍的症状）理解有限。在本工作中，我们利用当代测量理论来解码GPT-4和GPT-5如何相互关联抑郁症状，从而解释LLM如何应用其所学知识，并为临床应用提供信息。我们发现，GPT-4（a）与标准工具和专家判断具有强收敛效度（r = 0.70 - 0.81），并且（b）在行为上将抑郁症状彼此关联（症状间相关 r = 0.23 - 0.78），这与已有的抑郁症文献一致；然而，它（c）低估了自杀意念与其他症状之间的关系，同时高估了精神运动症状；并且（d）提出了关于症状机制的新假设，例如，表明睡眠和疲劳受到其他抑郁症状的广泛影响，而无价值感/内疚感仅与抑郁情绪相关。GPT-5显示出与自我报告的收敛性略低，我们的机器行为分析通过症状-症状关系的转变使这一差异变得可解释。这些见解为理解语言模型的心理健康评估提供了实证基础，并展示了一种可推广的方法，适用于其他模型和障碍的可解释性。我们的发现可以指导关键利益相关者做出明智决策，以有效将这些技术定位在护理系统中。

英文摘要：

Use of large language models such as ChatGPT (GPT-4/GPT-5) for mental health support has grown rapidly, emerging as a promising route to assess and help people with mood disorders like depression. However, we have a limited understanding of these language models' schema of mental disorders, that is, how they internally associate and interpret symptoms of such disorders. In this work, we leveraged contemporary measurement theory to decode how GPT-4 and GPT-5 interrelate depressive symptoms, providing an explanation of how LLMs apply what they learn and informing clinical applications. We found that GPT-4 (a) had strong convergent validity with standard instruments and expert judgments $(r = 0.70 - 0.81)$, and (b) behaviorally linked depression symptoms with each other (symptom inter-correlates $r = 0.23 - 0.78$) in accordance with established literature on depression; however, it (c) underemphasized the relationship between $\textit{suicidality}$ and other symptoms while overemphasizing $\textit{psychomotor symptoms}$; and (d) suggested novel hypotheses of symptom mechanisms, for instance, indicating that $\textit{sleep}$ and $\textit{fatigue}$ are broadly influenced by other depressive symptoms, while $\textit{worthlessness/guilt}$ is only tied to $\textit{depressed mood}$. GPT-5 showed a slightly lower convergence with self-report, a difference our machine-behavior analysis makes interpretable through shifts in symptom-symptom relationships. These insights provide an empirical foundation for understanding language models' mental health assessments and demonstrate a generalizable approach for explainability in other models and disorders. Our findings can guide key stakeholders to make informed decisions for effectively situating these technologies in the care system.

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。