跳到正文
研踪PaperTrail
← 研踪 PaperTrail

arXiv:2611.19345 · v1

Information Limits of Replay-Free Recurrent State Migration

首次提交 2026-11-14 19:34
版本更新 2026-11-14 19:34
雷达观测 2026-11-17 12:06
arXiv ↗PDF ↗

摘要

When a recurrent model is replaced, its old state may not determine what the new model would compute from the same history. We characterize the information that must be retained before the replacement is known. For a time-invariant affine linear family, the minimum continuous causal memory for order-kk local migration equals the reachable dimension of a lifted coefficient system minus the information in the anchor state. A controllable dd-state model with scalar input needs at most kdkd additional real coordinates, even when arbitrary matrix replacement directions are revealed only after streaming. A single rank-one direction can already require all kdkd. Conditional singular values and auxiliary-bit bounds distinguish this local information law from recovery at a fixed tolerance. We also give invariant-subspace and future-output constructions. Controlled experiments compare the resulting representations. In a small language model, equal-memory cubic interpolation improves mean state fidelity over third-order coefficients but provides no meaningful predictive gain. An analytic history basis preserves useful adaptation more economically in a coupled identification task. With repeated slow poles, eight history coordinates meet normalized output error 10−410^{-4} across five fitted seeds despite substantial full-state error. Splitting the poles raises the smallest tested successful width to 64.

AI 论文解读

基于摘要 · 非全文审阅

此版本尚未生成 AI 解读。解读由后台任务生成并按论文版本保存,登录后共享已有结果。

版本差异

目前只收录了一个版本。收录下一版后,可以对照标题、作者、分类和摘要。

观测到的版本

相邻记录可能跳过中间版本;首次发现 v2 及以上也会标为版本更新。

v1 · 2026-11-14 19:34 · 最新观测版本

Information Limits of Replay-Free Recurrent State Migration

When a recurrent model is replaced, its old state may not determine what the new model would compute from the same history. We characterize the information that must be retained before the replacement is known. For a time-invariant affine linear family, the minimum continuous causal memory for order-kk local migration equals the reachable dimension of a lifted coefficient system minus the information in the anchor state. A controllable dd-state model with scalar input needs at most kdkd additional real coordinates, even when arbitrary matrix replacement directions are revealed only after streaming. A single rank-one direction can already require all kdkd. Conditional singular values and auxiliary-bit bounds distinguish this local information law from recovery at a fixed tolerance. We also give invariant-subspace and future-output constructions. Controlled experiments compare the resulting representations. In a small language model, equal-memory cubic interpolation improves mean state fidelity over third-order coefficients but provides no meaningful predictive gain. An analytic history basis preserves useful adaptation more economically in a coupled identification task. With repeated slow poles, eight history coordinates meet normalized output error 10−410^{-4} across five fitted seeds despite substantial full-state error. Splitting the poles raises the smallest tested successful width to 64.

查看此版本解读 →查看此版本原文 ↗