让大模型主动清理无用记忆，提升长期任务成功率
MeClear: Cooperative Game-Theoretic Attribution and Risk-Aware Memory Clearance for Long-Horizon LLM Agents
- 用合作博弈论评估记忆价值，识别有害信息
- 清除后任务恢复率达82.3%，比基线高25.5个百分点
- 适合长期对话、复杂推理类大模型应用
长周期大型语言模型（LLM）代理依赖外部记忆系统保存用户偏好和任务知识。传统检索机制仅优化语义匹配，常引入过时、误导或矛盾的信息。我们提出MeClear，一种任务驱动的记忆清理框架，通过合作归因识别具有负下游效用的记忆，并选择性抑制其在执行中的影响。MeClear结合留一法筛选与采样合作式谢尔普利值归因，将效用分布到相互作用的记忆中，有效解决单次移除评估失效的冗余冲突掩盖问题。基于归因排序，执行查询范围最小化清理策略，通过嵌套过滤验证清理后上下文的任务恢复能力，且不永久修改持久化记忆库。在十个长对话记忆池上的全面实验表明，MeClear实现目标召回率85.9%和整体任务恢复率82.3%，较留一法基线提升25.5个百分点。
原文摘要 · Abstract (English)
Long horizon Large Language Model (LLM) agents rely on external memory systems to preserve user preferences and task knowledge across extended interactions. Conventional retrieval mechanisms optimize semantic compatibility rather than downstream utility, frequently introducing outdated, misleading, or conflicting evidence into the active context. We present MeClear, a task conditioned memory clearance framework that identifies memories featuring negative downstream utility through cooperative attribution and selectively suppresses them from agent execution. MeClear combines Leave One Out screening with sampled cooperative Shapley attribution to distribute utility across interacting evidence, effectively resolving redundant conflict masking where single removal evaluations fail. Utilizing attribution rankings, MeClear executes a query scoped minimal clearance strategy over a nested filtration, verifying task recovery on the cleared context without permanently altering the persistent memory bank. Comprehensive experimental evaluations across ten long dialogue memory pools demonstrate that MeClear achieves a target recall of 85.9% and an overall task recovery rate of 82.3%, representing a 25.5 percentage point improvement over Leave One Out (LOO) baselines.
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。