arXiv:2610.03020 · Agent 智能体
DyadMem: A Long-Term Memory Benchmark of How Agents Work with Users
DyadMem：关于 Agent 如何与用户协作的长期记忆基准
DyadMem: A Long-Term Memory Benchmark of How Agents Work with Users
- 类型：arxiv
- 标识：2610.03020
- 链接：http://arxiv.org/abs/2610.03020v1
- 主分类：agent
- 形态：benchmark
- TLDR：Long-term agents must remember not only what is true about a user, but also how a particular agent should work with that user as their shared history evolves. Existing benchmarks primarily supervise user facts and preferences or experience reusable across users, leaving this relationship-specific agent memory implicit. Additionally, most prior works measure the model solely with final-answer QA over long interaction histories, making the assessment still incomplete and unreliable. To this end, we introduce DyadMem with the proposed new definition User-conditioned Relational Agent Memory (URAM)
- 副分类：evaluation
- 待LLM分类：否
- 标题中文：DyadMem：关于 Agent 如何与用户协作的长期记忆基准
- TLDR中文：长期 Agent 不仅需要记住关于用户的真实信息，还需要记住随着共同历史演进，特定 Agent 应如何与该用户协作。现有基准主要监督用户事实和偏好，或监督跨用户可复用的经验，使这种关系特定的 Agent 记忆未被显式建模。此外，大多数先前工作仅通过长交互历史上的最终答案问答来评估模型，导致评估仍不完整且不可靠。为此，我们提出 DyadMem 及新定义——用户条件关系型 Agent 记忆（URAM）。
- 来源文件：
- /inbox/tom/_candidates/2026-10-05-agent-memory-tool-use-candidates.json