当持续学习转向记忆：对LLM智能体经验重用的研究
When Continual Learning Moves to Memory: A Study of Experience Reuse in LLM Agents
AI总结：
研究揭示了内存增强型LLM智能体在持续学习中经验重用的问题，发现记忆检索时旧经验与新经验的竞争加剧了持续学习的瓶颈，提出了(k,v)框架以解耦外部记忆的设计关键要素。
摘要：
内存增强型LLM智能体提供了一种持续学习的快捷方法：而不是更新模型参数，它们在外部记忆中积累经验，似乎规避了参数学习的稳定性-可塑性困境。我们证明这一挑战并未消失，而是在内存层面重新出现。在有限的上下文窗口下，旧经验和新经验在检索时竞争，将持续学习的瓶颈从参数更新转移到内存访问。为了研究这一现象，我们引入了一个(k,v)框架，以解耦外部记忆的两个基本设计轴：经验如何表示以及如何组织以供检索。在ALFWorld和BabyAI的连续任务实验中，我们发现抽象程序记忆比详细轨迹转移更可靠，而负迁移对困难案例产生不成比例的损害。此外，更细粒度的记忆组织并不总是有益：产生强正迁移的设计可能同时导致严重遗忘。这些结果表明，外部记忆并未解决持续学习的问题；它将问题转变为记忆表示和检索设计的问题。
英文摘要：
Memory-augmented LLM agents offer an appealing shortcut to continual learning: rather than updating model parameters, they accumulate experience in external memory, seemingly sidestepping the stability-plasticity dilemma of parametric learning. We show that this challenge does not disappear but resurfaces at the memory level. Under a limited context window, old and new experiences compete during retrieval, relocating the continual-learning bottleneck from parameter updates to memory access. To study this phenomenon, we introduce a (k,v) framework that disentangles two fundamental design axes of external memory: how experience is represented and how it is organized for retrieval. Across sequential-task experiments in ALFWorld and BabyAI, we find that abstract procedural memories transfer more reliably than detailed trajectories, while negative transfer disproportionately harms the hard cases. Moreover, finer-grained memory organization is not universally beneficial: designs that yield strong forward transfer can simultaneously induce severe forgetting. Together, these results reveal that external memory does not resolve the continual-learning problem; it reshapes it into a problem of memory representation and retrieval design.
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。
