Memory That Looks Forward: A Zero-Inference Prospective Term for Personal Memory Retrieval

面向未来的记忆：一种用于个人记忆检索的零推理前瞻项

Abstract: Retrieval over a personal memory store is retrospective: it surfaces what resembles the query, and it is blind to what the user has committed to do. We describe a prospective term for memory retrieval that costs no inference at query time.

摘要： 对个人记忆存储的检索通常是回顾性的：它呈现的是与查询相似的内容，却对用户已经承诺要做的事情视而不见。我们描述了一种用于记忆检索的前瞻项（prospective term），它在查询时无需任何推理成本。

Commitments are held in an explicit ledger as dated or trigger-conditioned entries; memory items linked to a firing entry receive a salience boost, blended multiplicatively into embedding-based retrieval so that relevance remains sovereign.

承诺以带日期或触发条件的条目形式保存在显式账本中；与触发条目相关联的记忆项会获得显著性提升，并以乘法方式融合到基于嵌入（embedding）的检索中，从而确保相关性始终占据主导地位。

On a synthetic prospective-memory task set modeled on TriggerBench’s published structure (48 blind-authored dialogues, 175 tasks), the term raised recall@5 on the hard stratum from 0.000 to 0.955 at the default blend weight and to 1.000 under a floor variant, with zero false boosts across 53 resolved-commitment tasks.

在一个模仿 TriggerBench 已发布结构（48 个盲写对话，175 个任务）的合成前瞻记忆任务集上，该项在默认融合权重下将困难层级的 recall@5 从 0.000 提升至 0.955，在底限变体下提升至 1.000，且在 53 个已解决的承诺任务中实现了零错误提升。

Blind authorship also produced a scope finding: only 17-29% of naturally phrased commitment-trigger pairs defeat embedding similarity, so the term matters on a real minority of cases and must do no harm on the rest, which it does not.

盲写实验还得出了一项关于适用范围的发现：只有 17-29% 的自然语言承诺-触发对能够超越嵌入相似度，因此该项仅在少数实际案例中起作用，且必须确保在其余案例中不产生负面影响，而实验证明它确实做到了这一点。

We position precomputed commitment linkage as the always-on floor of a layered design whose expansion layer is query-time prospection. Results are preliminary: the evaluation set is author-constructed, and evaluation on TriggerBench proper is committed follow-up work once its data is released.

我们将预计算的承诺链接定位为分层设计中始终开启的底层，其扩展层则是查询时的前瞻功能。目前结果尚属初步：评估集由作者构建，针对 TriggerBench 本身的评估将在其数据发布后作为后续工作进行。