KnowMe-Bench：为终身数字伴侣的人理解进行基准测试
KnowMe-Bench: Benchmarking Person Understanding for Lifelong Digital Companions
AI总结：
KnowMe-Bench通过长篇自传式叙述构建，评估模型在事实记忆、主观状态归因和原则推理方面的能力，揭示检索增强系统在时间相关解释和高层推理上的不足。
摘要：
现有长周期记忆基准主要使用多轮对话或合成用户历史，使检索性能成为人理解的不完美代理。我们提出了KnowMe-Bench，一个可公开释放的基准，基于长篇自传式叙述，其中行动、上下文和内心思想提供密集证据，用于推断稳定的动机和决策原则。KnowMe-Bench将每个叙述重构为具有回溯意识的时间锚定流，并通过涵盖事实记忆、主观状态归因和原则级推理的证据关联问题评估模型。在多样化的叙述来源中，检索增强系统主要提高事实准确性，但时间相关解释和高层推理仍存在错误，凸显了超越检索的记忆机制的必要性。我们的数据在https://github.com/QuantaAlpha/KnowMeBench。
英文摘要：
Existing long-horizon memory benchmarks mostly use multi-turn dialogues or synthetic user histories, which makes retrieval performance an imperfect proxy for person understanding. We present \BenchName, a publicly releasable benchmark built from long-form autobiographical narratives, where actions, context, and inner thoughts provide dense evidence for inferring stable motivations and decision principles. \BenchName~reconstructs each narrative into a flashback-aware, time-anchored stream and evaluates models with evidence-linked questions spanning factual recall, subjective state attribution, and principle-level reasoning. Across diverse narrative sources, retrieval-augmented systems mainly improve factual accuracy, while errors persist on temporally grounded explanations and higher-level inferences, highlighting the need for memory mechanisms beyond retrieval. Our data is in \href{KnowMeBench}{https://github.com/QuantaAlpha/KnowMeBench}.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
