发表机构
City University of Hong Kong; AgentWoods Inc.; Mohamed bin Zayed University of Artificial Intelligence（香港城市大学; AgentWoods公司; 穆罕默德·本·扎耶德人工智能大学）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
CreaMem通过场景感知划分和双视角编码解决长期记忆检索中的跨场景干扰与单视角局限，显著提升多跳推理问答准确率。
AI 中文摘要
长期记忆是个性化大语言模型智能体的核心能力。为支持这一能力，现有记忆系统使用各种标准（如主题片段或摘要层级）来组织信息。然而，我们发现了这些设计的两个主要局限。首先，它们缺乏场景感知：来自无关生活场景的记忆共享同一检索空间，这扩大了搜索空间并引入了跨场景干扰。其次，它们从单一视角编码每条记忆，使得难以检索同一事件的互补视角。在本文中，我们提出了CreaMem架构，该架构通过将记忆划分为多个生活场景记忆来实现场景感知的记忆组织，以减少检索时的跨场景干扰。为超越单一视角并实现跨记忆协同，每条记忆条目在每类记忆内均从情景和特质两个视角进行双重编码。我们进一步设计了一种检索时的每记忆平衡采样策略。在两个长期记忆基准上的大量实验表明，CreaMem在所有评估指标上均提升了问答准确率，尤其在多跳推理性能上取得了显著提升，验证了场景感知划分和跨记忆协同的有效性。为增强可复现性，我们在公开的GitHub仓库中发布了代码。
英文摘要
Long-term memory is a core capability for personalized LLM agents. To support it, existing memory systems organize information using various criteria such as topic segments or summary hierarchies. However, we identify two major limitations in these designs. First, they lack scene awareness: memories from unrelated life scenes share the same retrieval space, which inflates the search space and introduces cross-scene interference. Second, they encode each memory from a single perspective, making it difficult to retrieve complementary views of the same event. In this paper, we propose the CreaMem architecture, which enables scene-aware memory organization by partitioning memory into several Life Scene Memories to reduce cross-scene interference at retrieval. To go beyond the single perspective and achieve cross-memory synergy, entries are dual-coded from both episodic and trait-based perspectives within each memory. We further devise a permemory balanced sampling strategy at retrieval time. Extensive experiments on two long-term memory benchmarks show that CreaMem improves QA accuracy across all evaluation metrics, with particularly large gains on multi-hop reasoning performance, validating scene-aware partitioning and cross-memory synergy. To enhance reproducibility, we release our code in a public GitHub repository.
CommentsAccepted as EMNLP'26 Findings
