当智能体“误忆”时：探索基于大语言模型的多智能体系统中的曼德拉效应
When Agents "Misremember" Collectively: Exploring the Mandela Effect in LLM-based Multi-Agent Systems
AI总结：
本文研究了基于大语言模型的多智能体系统中的曼德拉效应，提出MANBENCH基准并提出缓解策略，实现74.40%的效应减少。
摘要：
近年来，大语言模型（LLMs）的进步显著增强了协作多智能体系统的能力，使它们能够解决复杂挑战。然而，在这些多智能体系统中，智能体对集体认知偏见的易感性仍是一个未被充分研究的问题。一个引人注目的例子是曼德拉效应，即群体因社会影响和内部化虚假信息而集体误记过去事件的现象。这种脆弱性限制了我们对多智能体系统中记忆偏见的理解，并引发了关于虚假信息潜在传播的伦理问题。在本文中，我们对基于大语言模型的多智能体系统中的曼德拉效应进行了全面研究，重点研究其存在性、成因因素和缓解策略。我们提出了MANBENCH，一种新的基准，旨在评估四种常见的易受曼德拉效应影响的任务类型，使用五种交互协议，这些协议在智能体角色和记忆时间尺度上有所不同。我们评估了由几种LLMs驱动的智能体在MANBENCH上的表现，以量化曼德拉效应并分析不同因素如何影响它。此外，我们提出了缓解这种效应的策略，包括提示层面的防御（例如认知锚定和来源审查）和模型层面的对齐防御，实现了与基线相比平均74.40%的曼德拉效应减少。我们的发现为开发更具韧性和伦理对齐的协作多智能体系统提供了有价值的见解。代码和数据集可在https://github.com/bluedream02/Mandela-Effect获取。
英文摘要：
Recent advancements in large language models (LLMs) have significantly enhanced the capabilities of collaborative multi-agent systems, enabling them to address complex challenges. However, within these multi-agent systems, the susceptibility of agents to collective cognitive biases remains an underexplored issue. A compelling example is the Mandela effect, a phenomenon where groups collectively misremember past events as a result of false details reinforced through social influence and internalized misinformation. This vulnerability limits our understanding of memory bias in multi-agent systems and raises ethical concerns about the potential spread of misinformation. In this paper, we conduct a comprehensive study on the Mandela effect in LLM-based multi-agent systems, focusing on its existence, causing factors, and mitigation strategies. We propose MANBENCH, a novel benchmark designed to evaluate agent behaviors across four common task types that are susceptible to the Mandela effect, using five interaction protocols that vary in agent roles and memory timescales. We evaluate agents powered by several LLMs on MANBENCH to quantify the Mandela effect and analyze how different factors affect it. Moreover, we propose strategies to mitigate this effect, including prompt-level defenses (e.g., cognitive anchoring and source scrutiny) and model-level alignment-based defense, achieving an average 74.40% reduction in the Mandela effect compared to the baseline. Our findings provide valuable insights for developing more resilient and ethically aligned collaborative multi-agent systems. Code and dataset are available at https://github.com/bluedream02/Mandela-Effect.
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。
