发表机构
Called It Inc.（Called It公司）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
本研究揭示时间记忆系统在旧语句重读与真实回退间产生歧义，提出拒绝重新激活已淘汰值的守卫方法，在重读条件下将准确率从10.8%恢复至97.7%，并强调区分观测与变更证据的设计启示。
AI 中文摘要
记忆系统可以淘汰一个过时的值，之后仅仅因为相同的旧语句再次出现而恢复该值。对旧来源的重读与真正的回退可能产生相同的观测值序列，却需要相反的当前答案。我们基于软件修复中提取的130个原子转换来研究这种歧义。在普通转换条件下，基于身份的时间记忆在字面过时值代理下达到了98.5%的模型判定准确率，且观测错误为零。附加旧语句的逐字重读后，准确率降至10.8%，过时值率升至88.5%。一个拒绝重新激活先前已淘汰值的守卫，在这种构造的重读条件下将准确率恢复至97.7%，并将过时值率降至0.8%。该守卫无法在没有额外变更来源信息的情况下识别合法的回退。两项辅助研究探讨了向答案模型暴露已淘汰历史以及为变更行为提供当前来源。一项基于707个软件修复的探索性提取研究提供了范围背景，而非普遍的覆盖率估计。设计启示在于区分对值的观测与值已变更的证据。所选输入、仅聚合的答案记录、相关族判定器以及故障后守卫评估将结论限制在所保留的实验范围内。
英文摘要
A memory system can retire an obsolete value and later restore it merely because the same old statement appears again. A re-read of an old source and a genuine revert can produce the same observed sequence of values while requiring opposite current answers. We study this ambiguity on 130 extractor-selected atomic transitions derived from software fixes. In the ordinary transition condition, identity-based temporal memory reaches 98.5% model-judged accuracy with zero observed errors under a literal stale-value proxy. Appending a verbatim re-read of the old statement reduces accuracy to 10.8% and raises the stale-value rate to 88.5%. A guard that refuses to reactivate a previously retired value restores accuracy to 97.7% and reduces that rate to 0.8% in this constructed re-read condition. The guard cannot also recognize a legitimate revert without additional change provenance. Two supporting studies examine exposing retired history to the answer model and supplying current source for changed behavior. An exploratory extraction study over 707 software fixes provides scope context, not a universal coverage estimate. The design implication is to distinguish an observation of a value from evidence that the value changed. Selected inputs, aggregate-only answer records, related-family judges and a post-failure guard evaluation limit the conclusions to the retained experiments.
Comments12 pages, 1 figure. Ancillary files contain retained aggregate evidence, derived scenario and annotation exports, reference code, and an offline verifier
