大型语言模型在人机交互中的心智理论能力：一种幻觉？
Theory of Mind abilities of Large Language Models in Human-Robot Interaction : An Illusion?
AI总结：
本文研究大型语言模型在人机交互中是否真正具备心智理论能力，通过感知行为识别任务和扰动测试揭示其高分表现源于对上下文扰动的脆弱性，而非真正具备心智理论。
摘要：
大型语言模型在各种自然语言和生成任务中展现出了卓越的生成能力。然而，可能的拟人化倾向以及对失败案例的宽容态度，推动了关于大型语言模型涌现能力的讨论，尤其是关于大型语言模型中心智理论（Theory of Mind, ToM）能力的讨论。虽然已有多种错误信念测试用于验证推断和维持另一个实体心理模型的能力，但我们研究的是心智理论能力的一个特殊应用场景，该场景具有更高的风险并可能带来不可逆的后果：人机交互（Human Robot Interaction）。在这项工作中，我们探索了感知行为识别（Perceived Behavior Recognition）任务，在该任务中，机器人使用大型语言模型（LLM）以类似于人类观察者的方式来评估机器人所生成的行为。我们重点关注四种行为类型，即——可解释的（explicable）、易读的（legible）、可预测的（predictable）和混淆性的（obfuscatory）行为，这些行为已被广泛用于合成可解释的机器人行为。因此，大型语言模型的目标是充当智能体的人类代理，并回答某一智能体行为将如何被回路中的人类所感知，例如“给定机器人的行为 X，人类观察者是否会认为它是可解释的？”。我们进行了一项人类受试者研究，以验证用户在五个领域的精心设计情境（机器人设置和计划）中能够正确回答此类问题。对信念测试的初步分析得出了极其积极的结果，这抬高了人们对大型语言模型具备心智理论能力的期望。随后，我们提出并执行了一套扰动测试，这些测试打破了这种幻觉，即不一致信念（Inconsistent Belief）、无信息上下文（Uninformative Context）和确信度测试（Conviction Test）。我们得出结论：大型语言模型在普通提示上的高分展示了其在人机交互环境中的潜在用途，但要具备心智理论，需要对上下文中琐碎或不相关的扰动保持不变性，而大型语言模型缺乏这种不变性。
英文摘要：
Large Language Models have shown exceptional generative abilities in various natural language and generation tasks. However, possible anthropomorphization and leniency towards failure cases have propelled discussions on emergent abilities of Large Language Models especially on Theory of Mind (ToM) abilities in Large Language Models. While several false-belief tests exists to verify the ability to infer and maintain mental models of another entity, we study a special application of ToM abilities that has higher stakes and possibly irreversible consequences : Human Robot Interaction. In this work, we explore the task of Perceived Behavior Recognition, where a robot employs a Large Language Model (LLM) to assess the robot's generated behavior in a manner similar to human observer. We focus on four behavior types, namely - explicable, legible, predictable, and obfuscatory behavior which have been extensively used to synthesize interpretable robot behaviors. The LLMs goal is, therefore to be a human proxy to the agent, and to answer how a certain agent behavior would be perceived by the human in the loop, for example "Given a robot's behavior X, would the human observer find it explicable?". We conduct a human subject study to verify that the users are able to correctly answer such a question in the curated situations (robot setting and plan) across five domains. A first analysis of the belief test yields extremely positive results inflating ones expectations of LLMs possessing ToM abilities. We then propose and perform a suite of perturbation tests which breaks this illusion, i.e. Inconsistent Belief, Uninformative Context and Conviction Test. We conclude that, the high score of LLMs on vanilla prompts showcases its potential use in HRI settings, however to possess ToM demands invariance to trivial or irrelevant perturbations in the context which LLMs lack.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
