Large language model (LLM) agents increasingly rely on accumulated memory to solve long-horizon decision-making tasks. However, most existing approaches store memory in fixed representations and reuse it at a single or implicit level of abstraction, which limits generalization and often leads to negative transfer when distribution shift. This paper proposes the Meta-Cognitive Memory Abstraction method (MCMA), which treats memory abstraction as a learnable cognitive skill rather than a fixed design choice. MCMA decouples task execution from memory management by combining a frozen task model with a learned memory copilot. The memory copilot is trained using direct preference optimization, it determines how memories should be structured, abstracted, and reused. Memories are further organized into a hierarchy of abstraction levels, enabling selective reuse based on task similarity. When no memory is transferable, MCMA transfers the ability to abstract and manage memory by transferring the memory copilot. Experiments on ALFWorld, ScienceWorld, and BabyAI demonstrate substantial improvements in performance, out-of-distribution generalization, and cross-task transfer over several baselines.

翻译：大语言模型（LLM）智能体日益依赖累积记忆来解决长视野决策任务。然而，现有方法大多将记忆存储为固定表示，并在单一或隐式的抽象层次上复用，这限制了泛化能力，且在分布偏移时常常导致负迁移。本文提出元认知记忆抽象方法（MCMA），该方法将记忆抽象视为一种可学习的认知技能，而非固定的设计选择。MCMA通过将冻结的任务模型与一个习得的记忆协处理器相结合，实现了任务执行与记忆管理的解耦。该记忆协处理器通过直接偏好优化进行训练，负责决定记忆应如何结构化、抽象化及复用。记忆进一步被组织成多层次的抽象层级，从而能够根据任务相似性进行选择性复用。当没有可迁移的记忆时，MCMA通过迁移记忆协处理器来迁移抽象与管理记忆的能力。在ALFWorld、ScienceWorld和BabyAI上的实验表明，相较于多种基线方法，MCMA在性能、分布外泛化及跨任务迁移方面均取得了显著提升。