大型语言模型在道德判断中的稳定性
1Department of Computer Science, Language Science and Technology, Saarland University, Saarbrücken, Germany.
Royal Society open science
|April 24, 2025
概括
关于大语言模型 (LLM) 和道德的先前研究是不可靠的. 我们的研究表明,LLM的道德判断对任务设计和提示措辞敏感,使以前的结论无效.
科学领域:
- 人工智能的人工智能
- 伦理学 伦理学 伦理学
- 自然语言处理自然语言处理.
背景情况:
- 越来越多的人对大语言模型 (LLM) 了解道德感兴趣.
- 之前的研究已经在道德判断任务上测试了LLM,这表明了类似人类的对齐.
- 有关用于评估LLM道德推理的方法存在担忧.
研究的目的:
- 批判性地评估LLM道德判断的先前研究的方法和结果.
- 调查任务呈现和快速制定对LLM响应的影响.
- 为更强大的未来研究设计提供建议.
主要方法:
- 分析现有的方法来从LLMs中获得道德判断.
- 测试LLM响应对任务选项呈现变化的敏感性.
- 检查微妙的提示措辞变化对LLM道德判断的影响.
主要成果:
- 之前的LLM道德判断研究被陈述偏见所混.
- 在LLM的响应中,对微小的快速配方变化有很高的敏感性.
- 从LLM道德对齐的先前工作中得出的结论没有得到支持.
结论:
- 评估LLM道德判断的现有方法缺乏有效性.
- 伦理理学士道德推理非常容易受到实验设计工件的影响.
- 未来的研究需要更严格和有偏见的方法来准确评估LLM的道德能力.
相关概念视频
Language and Cognition
291
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.
291
Confirmation Biases
5.4K
The confirmation bias is the tendency to focus on information that confirms our existing beliefs and ignore information that is inconsistent with our expectations. For example, if you think that your professor is not very nice, you notice all of the instances of rude behavior exhibited by the professor while ignoring the countless pleasant interactions he is involved in on a daily basis. Have you ever fallen prey to the confirmation bias, either as the source or target of such bias?
5.4K
Fundamental Attribution Error
12.8K
According to some social psychologists, people tend to overemphasize internal factors as explanations—or attributions—for the behavior of other people. They tend to assume that the behavior of another person is a trait of that person, and to underestimate the power of the situation on the behavior of others. They tend to fail to recognize when the behavior of another is due to situational variables, and thus to the person’s state. This erroneous assumption is...
12.8K
Kohlberg's Theory of Moral Development
25
Kohlberg's theory of moral development uses the Heinz dilemma — a thought experiment in which a man, Heinz, must decide whether to steal an unaffordable drug to save his dying wife — to illustrate the evolution of moral reasoning. This framework, divided into three levels with two stages, highlights how individuals' understanding of right and wrong becomes increasingly complex.
Pre-Conventional Level
At the pre-conventional level, morality is primarily driven by personal...
Pre-Conventional Level
At the pre-conventional level, morality is primarily driven by personal...
25
Language Development
280
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...
280
Lazarus's Cognitive Appraisal Theory
75
Cognitive psychologist Richard Lazarus proposed the cognitive-mediational theory of emotions, which emphasizes how individuals' assessments of stressors significantly affect their experience of stress. According to Lazarus, the stress response is determined by a two-step appraisal process: primary appraisal and secondary appraisal. These cognitive appraisals help individuals evaluate the potential impact of a stressor and determine the adequacy of their coping resources.
Primary Appraisal:...
Primary Appraisal:...
75


