PromptSense：具备情感感知与动态提示词的心理健康聊天机器人框架
PromptSense: Sentiment-Aware Mental Health Chatbot with Dynamic Prompting
AI 导读
一项 arXiv 研究提出 PromptSense 框架，在模型推理前先进行情感与抑郁分析，并通过规则过滤机制动态构建提示词，以提升 LLM 心理健康聊天机器人回复的共情性、上下文相关性与风险意识。
正文 · 原文
Abstract:Current Large Language Models (LLMs)-based mental health chatbots often lack sufficient empathy, contextual awareness, and safety mechanisms, potentially leading to ineffective or harmful interactions. This paper proposes a novel framework that integrates sentiment and depression analysis with a rule-based filtering mechanism prior to model inference. By dynamically constructing prompts based on emotional and risk-related insights, the framework aims to significantly improve the contextual relevance, empathy, and risk-aware of model-generated responses. We evaluated four model strategies and two prompt engineered variants across qualitative metrics such as empathy, helpfulness, clarity, relevance, safety, as well as toxicity assessments using the Perspective API. Experimental results indicate that the prompt engineering approaches outperforms DSFT and RLAIF by 1.8\% and 1.1\% respectively.
| Subjects: | Artificial Intelligence (cs.AI) |
| Cite as: | arXiv:2610.38044 [cs.AI] |
| (or arXiv:2610.38044v1 [cs.AI] for this version) | |
| https://doi.org/10.48550/arXiv.2610.38044 arXiv-issued DOI via DataCite |
Submission history
From: Jianlong Zhou [view email]
[v1]
Tue, 27 Oct 2026 22:57:49 UTC (2,360 KB)
来源：arXiv cs.AI · arxiv.org