BayesAgent: 通过 verbalized 概率图模型实现不确定性下的贝叶斯代理推理
BayesAgent: Bayesian Agentic Reasoning Under Uncertainty via Verbalized Probabilistic Graphical Modeling
AI总结：
BayesAgent 通过 verbalized 概率图模型实现不确定性下的贝叶斯代理推理，提升置信度校准和文本生成质量。
摘要：
人类认知在超越感官输入并形成潜在表示方面表现出色，这些表示结构了我们对世界的理解。尽管大型语言模型（LLM）代理展示了涌现的推理和决策能力，但它们缺乏一个系统框架来捕捉潜在结构并建模不确定性。在本工作中，我们首次探索如何将 LLM 代理与概率图模型（PGMs）结合起来，以解决不确定性下的代理推理问题。为此，我们引入了 verbalized 概率图模型（vPGM），一种贝叶斯代理框架，它通过自然语言指导 LLM 代理遵循 PGM 的关键原则，并通过数值贝叶斯推断细化由此产生的后验分布。与许多传统概率方法需要大量领域专业知识不同，vPGM 避免了专家驱动的模型设计，使其适用于假设有限的场景。我们评估了我们的模型在多个代理推理任务上的表现，包括封闭式和开放式任务。我们的结果表明，该模型有效提高了置信度校准和文本生成质量。
英文摘要：
Human cognition excels at transcending sensory input and forming latent representations that structure our understanding of the world. While Large Language Model (LLM) agents demonstrate emergent reasoning and decision-making abilities, they lack a principled framework for capturing latent structures and modeling uncertainty. In this work, we explore for the first time how to bridge LLM agents with probabilistic graphical models (PGMs) to address agentic reasoning under uncertainty. To this end, we introduce Verbalized Probabilistic Graphical Modeling (vPGM), a Bayesian agentic framework that (i) guides LLM agents in following key principles of PGMs through natural language and (ii) refines the resulting posterior distributions via numerical Bayesian inference. Unlike many traditional probabilistic methods requiring substantial domain expertise, vPGM bypasses expert-driven model design, making it well-suited for scenarios with limited assumptions. We evaluated our model on several agentic reasoning tasks, both close-ended and open-ended. Our results indicate that the model effectively enhances confidence calibration and text generation quality.
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。
