arXiv cs.CL· Chenhan Lyu, Yutong Song, Pengfei Zhang, Amir M. Rahmani·· 13 小时前AI 评分31
面向心理健康聊天机器人的领域特定 Constitutional AI 安全增强方法
Domain-Specific Constitutional AI: Enhancing Safety in LLM-Powered Mental Health Chatbots
AI 导读
研究者提出将 Constitutional AI 训练与领域特定的心理健康原则结合，用于提升 LLM 心理健康聊天机器人的安全性。该方法针对通用安全措施难以覆盖的危机干预准确性、治疗指南遵循、资源受限场景扩展性及细微对话中的偏差与痛苦信号识别等问题。
正文 · 原文
Abstract:Mental health applications have emerged as a critical area in computational health, driven by rising global rates of mental illness, the integration of AI in psychological care, and the need for scalable solutions in underserved communities. These include therapy chatbots, crisis detection, and wellness platforms handling sensitive data, requiring specialized AI safety beyond general safeguards due to emotional vulnerability, risks like misdiagnosis or symptom exacerbation, and precise management of vulnerable states to avoid severe outcomes such as self-harm or loss of trust. Despite AI safety advances, general safeguards inadequately address mental health-specific challenges, including crisis intervention accuracy to avert escalations, therapeutic guideline adherence to prevent misinformation, scale limitations in resource-constrained settings, and adaptation to nuanced dialogues where generics may introduce biases or miss distress signals. We introduce an approach to apply Constitutional AI training with domain-specific mental health principles for safe, domain-adapted CAI systems in computational mental health applications.
| Comments: | Accepted to 2025 IEEE 21st International Conference on Body Sensor Networks (BSN) |
| Subjects: | Artificial Intelligence (cs.AI); Computation and Language (cs.CL); Machine Learning (cs.LG) |
| ACM classes: | I.2.7; J.3; K.4.1 |
| Cite as: | arXiv:2509.16444 [cs.AI] |
| (or arXiv:2509.16444v2 [cs.AI] for this version) | |
| https://doi.org/10.48550/arXiv.2509.16444 arXiv-issued DOI via DataCite |
|
| Journal reference: | 2025 IEEE 21st International Conference on Body Sensor Networks (BSN), pp. 1-4 |
| Related DOI: | https://doi.org/10.1109/BSN66969.2025.11337405
DOI(s) linking to related resources |
Submission history
From: Chenhan Lyu [view email]
[v1]
Fri, 19 Sep 2025 21:46:47 UTC (704 KB)
[v2]
Tue, 20 Jan 2026 03:08:53 UTC (702 KB)
来源：arXiv cs.CL · arxiv.org