BaRe-Mem：用于稳健且自适应智能体咨询的贝叶斯可靠性记忆
BaRe-Mem: Bayesian Reliability Memory for Robust and Adaptive Agent Consultation
浏览论文内容
中文总结 AI 辅助
BaRe-Mem提出在线贝叶斯可靠性记忆，通过估计顾问可靠性并调节其影响，在咨询与自主推理间自适应选择，提升多智能体系统对误导信息的稳健性，并扩展到工人分配。
中文摘要 AI 辅助
在多智能体系统中，由于顾问能力在不同任务间存在差异，且误导性信息可能使咨询比自主推理更糟糕，因此可靠的咨询具有挑战性。我们提出BaRe-Mem，一种用于多智能体咨询的在线贝叶斯可靠性记忆。它基于中心模型的内部信念表示来估计顾问可靠性，并根据历史交互更新这些估计。这些估计调节顾问响应的影响，并指导在咨询与自主推理之间的选择。在九个基准和六个中心模型上，BaRe-Mem比辩论和多数投票对误导性顾问信息更稳健。在更具挑战性的任务上，在所有测试的误导水平下，它都保持在自主推理之上。此外，我们将BaRe-Mem机制扩展到智能体团队中的工人分配。在MuSiQue基准上，BaRe-Mem相比按历史成功次数路由，提高了任务完成率，并更早识别出有能力的工人。
英文摘要
In multi-agent systems, reliable consultation is challenging because advisor capabilities vary across tasks, and misleading information can make consultation worse than autonomous reasoning. We introduce BaRe-Mem, an online Bayesian reliability memory for multi-agent consultation. It estimates advisor reliability based on the central model's internal belief representations and updates these estimates from historical interactions. These estimates modulate the influence of advisor responses and guide the choice between consultation and autonomous reasoning. Across nine benchmarks and six central models, BaRe-Mem is more robust to misleading advisor information than debate and majority voting. On the more challenging tasks, it remains above autonomous reasoning across all tested misleading levels. Moreover, we extend the BaRe-Mem mechanism to worker allocation in agent teams. On the MuSiQue benchmark, BaRe-Mem improves task completion over routing by historical success counts and identifies capable workers earlier.
发表机构
- DeCLaRe Lab, Nanyang Technological University（南洋理工大学DeCLaRe实验室）
- Peking University（北京大学）
机构由 AI 辅助整理，请以论文原文为准。
