对语言模型(LMs)中的心理状态推理进行研究有潜力为人类社会认知理论(如心理状态推理部分源于语言暴露的理论)提供信息,并加深我们对LMs自身的理解。然而,许多发表的LMs研究依赖于相对较小的封闭源LMs样本,限制了我们严格测试心理理论和评估LM能力的能力。在此,我们通过在41个开放权重模型(来自不同模型家族)上评估LM的心理状态推理行为,复制并扩展了已发表的虚假信念任务研究。我们发现,在34%的测试LM中表现出对隐含知识状态的敏感性;然而,与先前研究一致,没有一个完全“解释掉”人类的效果。较大的LMs显示出更高的敏感性,并且也表现出更高的心理计量预测能力。最后,我们利用LM的行为生成并测试了一个关于人类认知的新假设:当使用非事实性动词(“John thinks...”)提示知识状态时,人类和LMs都倾向于倾向于归因虚假信念,而不是当使用间接提示(“John looks in the...”)时。与知识状态的主要效应不同,其中人类的敏感性超过LMs,人类知识提示效应的幅度正好落在LM效应大小的分布范围内——这表明语言统计学在原则上可以解释后者,但不能解释前者在人类中。这些结果证明了使用更大的开放权重LMs样本来测试人类认知理论和评估LM能力的价值。
英文摘要:
Research on mental state reasoning in language models (LMs) has the potential to inform theories of human social cognition--such as the theory that mental state reasoning emerges in part from language exposure--and our understanding of LMs themselves. Yet much published work on LMs relies on a relatively small sample of closed-source LMs, limiting our ability to rigorously test psychological theories and evaluate LM capacities. Here, we replicate and extend published work on the false belief task by assessing LM mental state reasoning behavior across 41 open-weight models (from distinct model families). We find sensitivity to implied knowledge states in 34% of the LMs tested; however, consistent with prior work, none fully ``explain away'' the effect in humans. Larger LMs show increased sensitivity and also exhibit higher psychometric predictive power. Finally, we use LM behavior to generate and test a novel hypothesis about human cognition: both humans and LMs show a bias towards attributing false beliefs when knowledge states are cued using a non-factive verb (``John thinks...'') than when cued indirectly (``John looks in the...''). Unlike the primary effect of knowledge states, where human sensitivity exceeds that of LMs, the magnitude of the human knowledge cue effect falls squarely within the distribution of LM effect sizes-suggesting that distributional statistics of language can in principle account for the latter but not the former in humans. These results demonstrate the value of using larger samples of open-weight LMs to test theories of human cognition and evaluate LM capacities.
Shanghai Institute of AI for Education and School of Computer Science and Technology, East China Normal University(华东师范大学上海智能教育研究院与计算机科学与技术学院);School of Data Science, Fudan University(复旦大学数据科学学院);Shanghai Key Laboratory of Data Science, School of Computer Science, Fudan University(复旦大学计算机科学学院上海市数据科学重点实验室)
Dissecting the Ullman Variations with a SCALPEL: Why do LLMs fail at Trivial Alterations to the False Belief Task?
Department of Psychology, 1202 W. Johnson Street Madison, WI 53706 USA(心理学系);Department of Educational Psychology, 1025 W. Johnson Street Madison, WI 53706 USA(教育心理学系)
Imperial College London(帝国理工学院);King’s College London(伦敦国王学院);University of Cambridge(剑桥大学);University of Oxford(牛津大学);University of Amsterdam(阿姆斯特丹大学)
Textual Unlearning Gives a False Sense of Unlearning
The State Key Laboratory of Blockchain and Data Security, Zhejiang University, P. R. China(区块链与数据安全国家重点实验室,浙江大学,中华人民共和国);School of Cyber Science and Technology, Zhejiang University, P. R. China(网络安全与技术学院,浙江大学,中华人民共和国);ETH Zurich, Switzerland(苏黎世联邦理工学院,瑞士)
Language Models as Critical Thinking Tools: A Case Study of Philosophers
Stanford University(斯坦福大学);University of Washington(华盛顿大学);Paul G. Allen School of Computer Science and Engineering, University of Washington(华盛顿大学保罗·G·艾伦计算机科学与工程学院)
Pushback in a Haystack: Language Models Respond to Factual Errors Ambiguously and Ineffectively
University of Washington(华盛顿大学);Allen Institute for AI(艾伦人工智能研究所);LMU Munich(慕尼黑大学);UC Berkeley(加州大学伯克利分校);Stanford University(斯坦福大学);University of Wisconsin-Madison(威斯康星大学麦迪逊分校)
Views Are My Own, but Also Yours: Benchmarking Theory of Mind Using Common Ground
Department of Computer Science(计算机科学系);Department of Linguistics(语言学系);Institute for Advanced Computational Science(高级计算科学研究院);Stony Brook University(石溪大学);University of Guilan(吉兰大学)
Reasoning Circuits in Language Models: A Mechanistic Interpretation of Syllogistic Inference
Department of Computer Science, University of Manchester, UK(英国曼彻斯特大学计算机科学系);Idiap Research Institute, Switzerland(瑞士Idiap研究所);School of Computer Science, University of Sheffield, UK(英国谢菲尔德大学计算机科学学院);National Biomarker Centre, CRUK-MI, University of Manchester, UK(英国曼彻斯特大学国家生物标志物中心)
Understanding Artificial Theory of Mind: Perturbed Tasks and Reasoning in Large Language Models
Bonn-Aachen International Center for Information Technology (b-it)(波恩-埃森国际信息科技中心);University of Bonn(波恩大学);Lamarr Institute for Machine Learning and Artificial Intelligence(拉马尔机器学习与人工智能研究所);Research Center Trustworthy Data Science and Security (RC-Trust)(可信数据科学与安全研究中心);University of Duisburg-Essen(埃森-杜伊斯堡大学)
本研究通过扰动任务和链式推理提示探讨大语言模型的理论 of mind能力,发现其鲁棒性受扰动影响显著,CoT提示虽提升整体表现,但对某些扰动类别却降低准确性。
Paul G. Allen School of Computer Science and Engineering, University of Washington(华盛顿大学保罗·G·艾伦计算机科学与工程学院);New York University(纽约大学);Allen Institute for Artificial Intelligence(艾伦人工智能研究所)