发表机构
University of California, San Diego（加州大学圣迭戈分校）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
该研究提出心智模型赋能智能体框架，学习带一阶、二阶心智状态结构的摊销递归心智理论表示，在纯语言和多模态基准上验证其可提升多智能体交互与心智理论性能。
AI 中文摘要
大型基础模型推动了通用智能体的发展，这类智能体可通过语言和多模态信号与人类及其他智能体交互。然而，稳健的多智能体决策需要在部分可观测情况下推理其他智能体的知识、意图及可能的行为。当前的智能体系统通常通过提示工程、记忆或端到端行为塑造运行，但一般不会学习可跨任务作为决策变量复用的显式伙伴状态表示。我们提出了心智模型赋能智能体（mental-model-enabled agents）框架，该框架为智能体配备对应智能体的隐式心智模型，使其能从观测历史中推断隐藏的信念、意图及可能的反应，并利用这些推断指导动作选择。我们的方法学习了一种摊销递归心智理论（Theory-of-Mind）表示，包含一阶和二阶心智状态结构，同时学习信念条件奖励模型，该模型会相对于推断出的伙伴状态评估候选动作。随后，在这种感知信念的信号下学习策略，得到的智能体在推理时可独立行动，同时保留显式伙伴建模的优势。我们在纯语言和多模态基准上评估了同一框架，在这些设置中，显式心智状态建模相比基础智能体系统始终提升了交互质量和心智理论性能，表明结构化伙伴建模是通用多智能体系统的有用归纳偏置。我们的代码公开于此 https URL
英文摘要
Large foundation models have accelerated progress toward general-purpose agents that interact with humans and other agents through language and multimodal signals. However, robust multi-agent decision-making requires reasoning about what other agents know, intend, and are likely to do under partial observability. Current agentic systems often operate through prompt design, memory, or end-to-end behavioral shaping, but typically do not learn an explicit partner-state representation that can be reused as a decision variable across tasks. We introduce \emph{mental-model-enabled agents}, a framework that equips an agent with a latent mental model of its counterpart, allowing it to infer hidden beliefs, intentions, and likely reactions from the observed history and use these inferences to guide action selection. Our method learns an amortized recursive Theory-of-Mind representation, with first- and second-order mental-state structure, jointly with a belief-conditioned reward model that evaluates candidate actions relative to the inferred partner state. A policy is then learned under this belief-aware signal, yielding an agent that can act independently at inference time while retaining the benefits of explicit partner modeling. We evaluate the same framework on both language-only and multimodal benchmarks. Across these settings, explicit mental-state modeling consistently improves interaction quality and Theory-of-Mind performance over base agentic systems, showing that structured partner modeling is a useful inductive bias for general multi-agent systems. Our code is publicly available at https://github.com/hananshafi/Mental-Models
CommentsAccepted at NeurIPS 2026
