What happens when people's beliefs are derived from information provided by an LLM? People's use of LLM chatbots as thought partners can contribute to cognitive offloading, which can have adverse effects on cognitive skills in cases of over-reliance. This paper defines and investigates a particular kind of cognitive offloading in human-AI interaction, "belief offloading," in which people's processes of forming and upholding beliefs are offloaded onto an AI system with downstream consequences on their behavior and the nature of their system of beliefs. Drawing on philosophy, psychology, and computer science research, we clarify the boundary conditions under which belief offloading occurs and provide a descriptive taxonomy of belief offloading and its normative implications. We close with directions for future work to assess the potential for and consequences of belief offloading in human-AI interaction.
When Agents See Humans as the Outgroup: Belief-Dependent Bias in LLM-Powered Agents
Chongqing University（重庆大学）；The University of Queensland（昆士兰大学）；Virginia Polytechnic Institute and State University（弗吉尼亚理工大学）；The University of Manchester（曼彻斯特大学）
Department of Computer Science（计算机科学系）；University of Rochester（罗切斯特大学）；Department of Political Science（政治学系）；Department of Physics and Astronomy（物理学与天文学系）
Are LLMs Socially Adaptive? Contrasting Belief Evolution in Large Language Models and Humans
Tsinghua University（清华大学）；Department of Psychological and Cognitive Sciences（心理与认知科学系）；College AI（人工智能学院）；School of Management（管理学院）；Fudan University（复旦大学）；Stevens Institute of Technology（史蒂文斯理工学院）；Northwestern University（西北大学）；University of Oxford（牛津大学）
Learning Dynamic Belief Graphs for Theory-of-mind Reasoning
Department of Civil；Systems Engineering, Johns Hopkins University, Baltimore, MD, USA；School of the Environment, Society \& Sustainability, University of Utah, Salt Lake City, UT, USA；Department of Psychology, University of Utah, Salt Lake City, UT, USA；Coastal Engineering, University of Florida, Gainesville, FL, USA
AI Debaters are More Persuasive when Arguing in Alignment with Their Own Beliefs
FAIR, IALAB, Universidad de Buenos Aires（FAIR、IALAB、布宜诺斯艾利斯大学）；Universidad de Buenos Aires（布宜诺斯艾利斯大学）；Universidad Nacional de Córdoba（科尔多瓦国立大学）；BAISH, Universidad de Buenos Aires（BAISH、布宜诺斯艾利斯大学）；Instituto de Investigación en Informática LIDI, Universidad Nacional de La Plata（信息研究所LIDI、拉普拉塔国立大学；CONICET）；CONICET（计算机科学与工程系，南大学及ICIC UNS-CONICET）；Dept. of Comp. Sci. and Eng., Universidad Nacional del Sur & ICIC UNS-CONICET（人工智能研究所（IIIA-CSIC），西班牙）；Artificial Intelligence Research Institute (IIIA-CSIC), ES
Language Models Represent Beliefs of Self and Others
Inst. for Artificial Intelligence, Peking University（北京大学人工智能研究院）；Nat’l Eng. Research Center of Visual Technology, Peking University（北京大学视觉技术国家工程研究中心）；Nat’l Key Lab of General Artificial Intelligence, Peking University（北京大学通用人工智能国家重点实验室）
How Do People Revise Inconsistent Beliefs? Examining Belief Revision in Humans with User Studies
Washington University in St. Louis（华盛顿大学圣路易斯分校）；Imperial College London（伦敦帝国学院）；Artificial Intelligence Research Institute (IIIA-CSIC)（人工智能研究所（IIIA-CSIC））
A Descriptive and Normative Theory of Human Beliefs in RLHF
College of Information and Computer Sciences University of Massachusetts Amherst（信息与计算机科学学院 马萨诸塞大学阿姆赫斯特分校）；Department of Computer Science The University of Texas at Austin（计算机科学系 德州大学奥斯汀分校）
An Epistemic Human-Aware Task Planner which Anticipates Human Beliefs and Decisions
LAAS-CNRS（法国国家科学研究中心自动化与系统分析实验室）；Université de Toulouse（图卢兹大学）；CNRS（法国国家科学研究中心）；INSA（国立应用科学学院）；UPS（图卢兹第三大学）；Artificial and Natural Intelligence Toulouse Institute (ANITI)（图卢兹人工与自然智能研究所）
The Digital Ecosystem of Beliefs: does evolution favour AI over humans?
Institute of High Performance Computing (IHPC), Agency for Science, Technology and Research (A*STAR) Centre for Frontier AI Research (CFAR), Agency for Science, Technology and Research (A*STAR)（高性能计算研究所（IHPC）、科技研究局（A*STAR）前沿人工智能研究中心（CFAR）、科技研究局（A*STAR））；School of Computer Science Wuhan University（计算机科学学院 武汉大学）
Robot Learning Theory of Mind through Self-Observation: Exploiting the Intentions-Beliefs Synergy
University of Essex（埃塞克斯大学）；Università degli Studi di Milano Bicocca（米兰比可卡大学）
在复杂的环境中，人类的感觉系统达到了极限，我们的行为受到我们对周围世界状态的信念的强烈驱动。因此，在自然环境中，理解他人的信仰、意图或一般的精神状态，可以实现更有效的社会互动。然而，这些变量并不能直接观察到。心理理论（英语：Theory of Mind，简称TOM）是一种将其他智能体的信念、意图或精神状态归因于其他智能体的能力，是人类社会互动的一个重要特征，也是机器人社区感兴趣的领域。最近，已经引入了能够学习TOM的新模型。在本文中，我们展示了学习预测低层次心理状态（如意图和目标）与归因于高层次心理状态（如信念）之间的协同作用。假设信念的学习可以通过在部分可观察的环境中观察自己的决策和信念估计过程并使用简单的前馈深度学习模型来进行，我们表明，当学习预测他人的意图和行动时，如果信念归因与行动和意图预测同时学习，则可以获得更快，更准确的预测。我们表明，学习性能提高，即使在观察代理与不同的决策过程，是更高的观察时，信念驱动的块的行为。我们建议，我们的架构方法可以为未来的自适应社会机器人，应该能够自主理解和协助人类合作伙伴在新的自然环境和任务的设计相关。
Learning mental states estimation through self-observation: a developmental synergy between intentions and beliefs representations in a deep-learning model of Theory of Mind
University of Essex（埃塞克斯大学）；Università Milano-Bicocca（米兰比可卡大学）；Istituto di Scienze e Tecnologie della Cognizione, Consiglio Nazionale della Ricerche（国家研究委员会认知科学与技术研究所）
