GateMem：多主共享内存代理中的内存管理基准测试
GateMem: Benchmarking Memory Governance in Multi-Principal Shared-Memory Agents
摘要
对于LLM智能体而言，内存性能测试大多基于单用户场景，而医院、工作场所、校园和家庭环境中使用的共享智能体则被忽视。在这些场景中，多个主体会向同一个内存池写入数据，并以不同的角色、范围和关系进行查询，因此内存质量需要相应的管理机制以及有效的信息检索能力。我们提出了GateMem这一用于多主体共享内存智能体的测试框架。GateMem能够综合评估以下方面的表现：合法的长距离请求处理能力、状态更新的处理效果、跨上下文授权边界的访问控制机制，以及智能体在明确删除操作后对信息的主动遗忘能力。该测试框架适用于医疗、办公、教育和家庭等多个领域，包含长篇幅的多方交互场景、逐步注入内存的机制、隐藏检查点、结构化评估方式以及针对信息泄露的标注功能。尽管存在多种基准模型和核心模型，但没有任何方法能够同时实现强大的性能、可靠的访问控制和可靠的信息遗忘机制。长上下文提示往往能在较高的token成本下获得最佳的治理效果，而基于检索和外部内存的方法虽然降低了成本，却仍可能导致未经授权的或已删除的信息泄露。这些结果表明，当前的内存智能体仍然无法满足可靠共享使用的要求。
English Abstract
Memory benchmarks for LLM agents largely assume single-user settings, leaving shared assistants for hospitals, workplaces, campuses, and households understudied. In these deployments, multiple principals write to a common memory pool and query it under different roles, scopes, and relationships, so memory quality requires governance as well as recall. We introduce GateMem, a benchmark for multi-principal shared-memory agents. GateMem jointly evaluates utility for legitimate long-horizon requests with state updates, access control across contextual authorization boundaries, and agent-facing active forgetting after explicit deletion requests. It spans medical, office, education, and household domains, with long-form multi-party episodes, incremental memory injection, hidden checkpoints, structured judging, and leak-target annotations. Across diverse baselines and backbone models, no method simultaneously achieves strong utility, robust access control, and reliable forgetting. Long-context prompting often yields the best governance score at high token cost, while retrieval-based and external-memory methods reduce cost yet still leak unauthorized or deleted information. These results show current memory agents remain far from reliable shared institutional deployment.