让AI智能选择记忆方式，提升长时间对话的连贯性。
Choosing How to Remember: Adaptive Memory Structures for LLM Agents
- 根据上下文动态选择不同记忆结构，不再固定使用一种方式。
- 在两个长程任务上分别提升9.18%和6.14%的表现。
- 适合需要长期记忆与复杂交互的AI代理系统开发。
记忆对大语言模型（LLM）驱动的智能体在长时间交互中保持行为一致性至关重要。然而，现有记忆系统存在两大缺陷：采用统一记忆结构且未将结构选择作为上下文自适应决策，难以应对多样化的交互模式，导致性能不佳。我们提出统一框架FluxMem，实现LLM智能体的自适应记忆组织。该框架赋予智能体多种互补的记忆结构，并基于交互级特征显式学习结构选择策略，利用下游响应质量与记忆利用率的离线监督信号。为支持稳健的长周期记忆演化，进一步引入三级记忆层级与基于贝塔混合模型的概率门控机制，实现分布感知的记忆融合，替代脆弱的相似性阈值。在两个长周期基准测试PERSONAMEM和LoCoMo上的实验表明，该方法分别实现平均9.18%和6.14%的性能提升。
原文摘要 · Abstract (English)
Memory is critical for enabling large language model (LLM) based agents to maintain coherent behavior over long-horizon interactions. However, existing agent memory systems suffer from two key gaps: they rely on a one-size-fits-all memory structure and do not model memory structure selection as a context-adaptive decision, limiting their ability to handle heterogeneous interaction patterns and resulting in suboptimal performance. We propose a unified framework, FluxMem, that enables adaptive memory organization for LLM agents. Our framework equips agents with multiple complementary memory structures. It explicitly learns to select among these structures based on interaction-level features, using offline supervision derived from downstream response quality and memory utilization. To support robust long-horizon memory evolution, we further introduce a three-level memory hierarchy and a Beta Mixture Model-based probabilistic gate for distribution-aware memory fusion, replacing brittle similarity thresholds. Experiments on two long-horizon benchmarks, PERSONAMEM and LoCoMo, demonstrate that our method achieves average improvements of 9.18% and 6.14%.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。