大型语言模型可能难以检测文化内嵌的杀-自杀风险
Cheng-Che Chen1, Justin A Chen2, Chih-Sung Liang3
1Department of Psychiatry, National Taiwan University Hospital Hsinchu Branch and Biomedical Park Hospital, Hsinchu, Taiwan; Department of Psychiatry, College of Medicine, National Taiwan University, Taipei, Taiwan.
Asian journal of psychiatry
|February 16, 2025
概括
大型语言模型 (LLM) 在检测台湾故事中的家庭暴力和自杀风险方面取得了不同程度的成功. 对于心理健康评估至关重要的文化细微差别仍然是人工智能解释的重大挑战.
科学领域:
- 人工智能的人工智能
- 心理健康 心理健康
- 社会文化研究 社会文化研究
背景情况:
- 大型语言模型 (LLM) 越来越多地用于各种分析任务.
- 评估心理健康风险需要细微的理解,包括文化背景.
- 之前的研究还没有广泛评估LLM在检测特定文化风险方面的能力.
研究的目的:
- 评估六个高级LLM在识别家庭暴力,自杀和菲利基德-自杀风险方面的表现.
- 评估LLM在台湾闪电小说中对文化特定线索的理解.
- 确定当前LLM在文化敏感心理健康风险检测方面的局限性.
主要方法:
- 六位法学士 (GPT-4o,GPT-o1,DeepSeek-R1,克劳德3.5索内特,索纳大,杰玛-2-2b) 分析了闪光小说"烧烤".
- 模型被要求解释风险,模拟不同的心理健康专业知识水平.
- 分析重点是检测家庭暴力,自杀和亲生杀人-自杀,并关注文化背景.
主要成果:
- 所有LLM检测到家庭暴力风险.
- GPT-o1,克劳德3.5索内特和索纳大用文化线索识别了自杀风险;其他人错过了它.
- 没有LLM理解母亲遗弃女儿的文化背景,这表明在认识非西方社会文化细微差别方面存在差距.
结论:
- 在检测家庭暴力等公开风险方面,LLM表现出部分有效性.
- 目前的LLM在解释文化中嵌入的自杀和菲利基德-自杀风险方面扎,特别是那些依赖于非西方背景的人.
- 需要进一步发展,以提高LLM的文化能力,以准确评估心理健康风险.
更多相关视频
08:53Integrating Computerized Linguistic and Social Network Analyses to Capture Addiction Recovery Capital in an Online Community
Published on: May 31, 2019
5.1K
09:09Foreign Accent and Forensic Speaker Identification in Voice Lineups: The Influence of Acoustic Features Based on Prosody
Published on: September 27, 2024
386
相关概念视频
01:27Language and Cognition
317
Language serves as a bridge between ideas and communication, influencing how individuals perceive and interact with the world. Psychologists have long debated whether language shapes thought or vice versa. This discussion gained grip with Edward Sapir and Benjamin Lee Whorf in the 1940s, who proposed that language determines thought, a concept known as linguistic determinism. They suggested that the vocabulary and structure of a language influence how its speakers think and perceive reality.
317
02:16Stereotype Content Model
13.9K
The Stereotype Content Model (SCM) was first proposed by Susan Fiske and her colleagues (Fiske, Cuddy, Glick & Xu, 2002; see also Fiske, 2012 and Fiske, 2017). The SCM specifies that when someone encounters a new group, they will stereotype them based on two metrics: warmth—or that group’s perceived intent, and how likely they are to provide help or inflict harm—and competence—or their ability to carry out that objective. Depending on the warmth-competence...
13.9K
01:22Language Development
308
Children master language quickly and with relative ease, supported by both biological predisposition and reinforcement. B. F. Skinner (1957) proposed that language is learned through reinforcement, while Noam Chomsky (1965) argued that language acquisition mechanisms are biologically determined.
The critical period for language acquisition suggests that the ability to acquire language is at its peak early in life. As people age, this proficiency decreases. Language development begins very...
The critical period for language acquisition suggests that the ability to acquire language is at its peak early in life. As people age, this proficiency decreases. Language development begins very...
308