用结构化记忆防遗忘错乱，让长时代理更懂信息从哪来。
Mitigating Provenance-Role Collapse in Long-Term Agents via Typed Memory Representation
- 将记忆拆成证据、线索、断言三类原子，分清来源与真假
- 在源追踪和时间定位任务上，性能超越现有方法23%以上
- 适合需要长期记忆与事实溯源的智能体系统
长期记忆对持续性大模型代理至关重要，但现有架构将历史交互存储为无结构的扁平文本，导致来源监控失效这一关键缺陷——即证明来源角色坍塌。为从架构层面解决这一认知漏洞，我们提出MemIR，一种带类型的记忆中间表示，将来源监控转化为结构性约束。MemIR将长期记忆写入具身原子，分离原始证据、检索线索与承载事实的断言，并限制事实授权仅限于有支持的断言原子。随后通过多路径原子投影与来源作用域利用，将异构检索结果转化为以断言为中心的候选包，并生成标准化的事实接口用于答案生成。在LoCoMo与BEAM-100K数据集上的实验表明，MemIR在需源追踪、时间定位及碎片证据聚合的任务中，持续优于现有记忆基线，尤其在复杂推理场景下提升显著。
原文摘要 · Abstract (English)
Long-term memory is essential for persistent LLM agents, yet prevailing architectures store historical interactions as unstructured, flat text. This unconstrained storage induces provenance-role collapse, a critical failure mode where agents suffer from source-monitoring errors. To resolve this cognitive vulnerability at the architectural level, we propose MemIR, a typed Memory Intermediate Representation that operationalizes source monitoring as a structural constraint. MemIR writes long-term memory into grounded atoms that separate raw evidence, retrieval cues, and truth-bearing claims, with factual authorization restricted to supported claim atoms. It then applies multi-route atomic projection and provenance-scoped utilization to transform heterogeneous retrieval hits into claim-centered candidate bundles and a normalized fact interface for answer generation. Experiments on LoCoMo and BEAM-100K demonstrate that MemIR consistently outperforms existing memory baselines, especially on tasks requiring source tracking, temporal grounding, and aggregation of fragmented evidence.
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。