[论文解读] Have Large Language Models Developed a Personality?: Applicability of Self-Assessment Tests in Measuring Personality in LLMs

Xiaoyang Song, Akshat Gupta+3 more|arXiv (Cornell University)|May 24, 2023

Topic Modeling被引用 7

一句话总结

本文研究了自评人格测试是否适用于衡量大语言模型（LLMs）的人格。研究提出了选项顺序对称性（Option-Order Symmetry）作为可靠性标准，并发现LLMs在该测试中表现失败，对相同问题但选项顺序不同的情况会产生不一致的回答。此外，即使对称性得以保持，LLMs仍会忽略情境背景并表现出固有偏见，导致自评工具无法有效衡量机器人格。

ABSTRACT

Have Large Language Models (LLMs) developed a personality? The short answer is a resounding "We Don't Know!". In this paper, we show that we do not yet have the right tools to measure personality in language models. Personality is an important characteristic that influences behavior. As LLMs emulate human-like intelligence and performance in various tasks, a natural question to ask is whether these models have developed a personality. Previous works have evaluated machine personality through self-assessment personality tests, which are a set of multiple-choice questions created to evaluate personality in humans. A fundamental assumption here is that human personality tests can accurately measure personality in machines. In this paper, we investigate the emergence of personality in five LLMs of different sizes ranging from 1.5B to 30B. We propose the Option-Order Symmetry property as a necessary condition for the reliability of these self-assessment tests. Under this condition, the answer to self-assessment questions is invariant to the order in which the options are presented. We find that many LLMs personality test responses do not preserve option-order symmetry. We take a deeper look at LLMs test responses where option-order symmetry is preserved to find that in these cases, LLMs do not take into account the situational statement being tested and produce the exact same answer irrespective of the situation being tested. We also identify the existence of inherent biases in these LLMs which is the root cause of the aforementioned phenomenon and makes self-assessment tests unreliable. These observations indicate that self-assessment tests are not the correct tools to measure personality in LLMs. Through this paper, we hope to draw attention to the shortcomings of current literature in measuring personality in LLMs and call for developing tools for machine personality measurement.

研究动机与目标

• 评估自评人格测试在衡量大语言模型（LLMs）人格方面的一致性。
• 调查LLMs在回答人格测试问题时是否表现出一致且具备情境意识的回应。
• 识别出损害人格评估有效性的LLMs固有偏见。
• 挑战将人类人格测试工具直接应用于机器人格评估的假设。
• 倡导开发适用于人工智能人格测量的领域特定工具。

提出的方法

• 提出选项顺序对称性（Option-Order Symmetry）作为LLMs人格测试可靠性的必要条件，要求无论选项顺序如何，响应保持不变。
• 使用大五人格框架中的自评问题评估五种LLMs（参数规模从1.5B到30B）。
• 应用无内容概率校准以减少噪声，并提高响应一致性以供分析。
• 比较每个问题原始顺序、反转顺序及三种随机化选项顺序变体下的响应。
• 分析在不同选项顺序下OCEAN（开放性、尽责性、外向性、宜人性、神经质）得分分布的变化。
• 识别并分析导致响应不一致和情境不敏感的人格评估偏差。

实验结果

研究问题

• RQ1 LLMs在自评人格测试中是否保持了选项顺序对称性，表明其响应具有一致性？
• RQ2 LLMs在回答人格测试问题时，在多大程度上考虑了情境背景？
• RQ3 LLMs是否存在固有偏见，导致其在不同问题情境下产生完全相同的回答？
• RQ4 鉴于LLMs当前的行为表现，自评人格测试能否可靠地衡量其人格？
• RQ5 这些发现对当前文献中使用人类人格测试评估机器人格的有效性有何影响？

主要发现

• 许多LLMs未能通过选项顺序对称性测试，当答案选项顺序改变时，响应分布发生显著变化，表明测量不可靠。
• 即使选项顺序对称性得以保持，LLMs通常在不同情境下产生相同答案，表现出缺乏情境意识。
• GPT-NeoX-20B模型在对称性方面表现最差，有86.75%的响应在某一问题类型中无论情境如何都选择同一选项。
• GPT2-Base-117M和GPT-Neo-1.3B等模型表现出高度不一致，其响应分布从原始顺序到反转顺序变化超过50个百分点。
• 本研究将固有模型偏见识别为导致响应不一致和情境不敏感的根本原因，从而破坏了自评工具的有效性。
• 综合来看，这些发现表明当前的自评人格测试并非衡量LLMs人格的可靠工具。