arXiv cs.CL· Bichen Wang, Yufeng Jin, Yixin Sun, Yanyan Zhao, Bing Qin·· 11 小时前AI 评分47
MindEth-Bench：评估中文语境下 LLM 心理健康支持的多轮伦理边界
When Support Crosses the Line: Evaluating Ethical Boundaries in LLM-Based Mental Health Support in the Chinese Context
AI 导读
研究者提出 MindEth-Bench，一个基于中国心理健康服务专业伦理标准、用 MCTS 系统探索压力测试交互轨迹的基准，用于评估 LLM 多轮心理健康对话中的伦理边界遵循情况。在九个 LLM 上的实验显示，初始回应得当的模型会随对话延长滑向伦理违规，且模型间与伦理维度间差异显著，说明单轮评估可能高估这类系统的伦理可靠性。
正文 · 原文
Abstract:Large language models (LLMs) are increasingly explored for mental health support, yet existing benchmarks primarily focus on supportive capabilities rather than whether models consistently adhere to professional ethical boundaries. This issue is particularly important in multi-turn interactions, where users may introduce new information or exert emotional pressure over time. We introduce MindEth-Bench, a benchmark for evaluating ethical boundary adherence in multi-turn mental health interactions. Grounded in professional ethical standards for mental health services in China, MindEth-Bench covers multiple dimensions of ethical conduct and uses Monte Carlo Tree Search (MCTS) to systematically explore interaction trajectories that stress-test model boundaries. Experiments across nine LLMs show that models that initially respond appropriately can shift toward ethical violations over extended interactions, with substantial variation across models and ethical dimensions. These results show that single-turn evaluations may overestimate the ethical reliability of LLM-based mental health support systems and highlight the need to evaluate ethical behavior throughout multi-turn interaction.
| Subjects: | Computation and Language (cs.CL) |
| Cite as: | arXiv:2610.38452 [cs.CL] |
| (or arXiv:2610.38452v2 [cs.CL] for this version) | |
| https://doi.org/10.48550/arXiv.2610.38452 arXiv-issued DOI via DataCite |
Submission history
From: Yufeng Jin [view email]
[v1]
Wed, 28 Oct 2026 08:30:26 UTC (756 KB)
[v2]
Fri, 30 Oct 2026 12:33:29 UTC (753 KB)
来源：arXiv cs.CL · arxiv.org