测试大模型记忆系统的五大故障模式，揭示其设计缺陷。
MemFail: Stress-Testing Failure Modes of LLM Memory Systems
- 将记忆系统拆解为摘要、存储、检索三步，逐项诊断故障。
- 构建五套对抗数据集，验证四种主流记忆系统在不同环节的失效表现。
- 适合研究大模型可靠性、记忆机制与系统设计的开发者参考。
大型语言模型（LLM）代理越来越多地依赖外部记忆系统以保持长周期交互中的一致性，但针对这些系统具体故障模式和设计选择的实证研究仍十分有限。现有基准仅报告整体问答准确率，将记忆系统视为黑盒，无法定位错误答案的具体原因。本文提出MemFail，一个用于诊断现代LLM记忆系统故障模式的基准测试。我们首先将记忆系统形式化为三个典型操作——摘要、存储、检索，并识别每个操作可能引发的故障模式。基于这些假设，构建了覆盖四个任务的五套数据集，每套数据集均经过对抗设计，专门测试记忆系统某一特定操作的表现。利用这些数据集，评估了四种先进记忆系统在MemFail上的表现，展示了如何通过该基准实证分析不同架构带来的权衡取舍。
原文摘要 · Abstract (English)
Large language model (LLM) agents increasingly rely on external memory systems to remain consistent across long-horizon interactions, but little empirical work has been done to understand the specific failure modes and design choices that these systems present. Existing benchmarks report aggregate question-answering accuracy and treat memory systems as black boxes, making it impossible to attribute an incorrect answer to a particular failure mode of the system. We introduce MemFail, a diagnostic benchmark that isolates the failure modes of modern LLM memory systems. We begin by formalizing memory systems as the composition of three canonical operations -- summarization, storage, and retrieval -- and identify the potential failure modes induced by each. Based on these hypothesized failure modes, we construct five datasets spanning four tasks, each adversarially designed to test a specific operation of a memory system. Using these datasets, we evaluate four state-of-the-art memory systems on MemFail and demonstrate how MemFail can be used to empirically understand the tradeoffs induced by differences in memory system architectures.
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。