Which Host Versions Should a Removable Branch Learn From?

A branch trained beside a continuing host should be removable without changing the host's training. Optimization-Recoverable Augmentation () enforces this through complete-state isolation, with conditional guarantees and Qwen/BERT recovery tests. Which host versions should the branch observe before deployment on the final host? In one-block Qwen2.5-0.5B, reader-progress observation ($P$) exceeds terminal-only training by $4.67/13.87\mathrm{pp}$ on GSM8K/SQuAD with only the version sequence changed. The advantage survives terminal learning-rate and checkpoint selection: $+5.25/+6.06\mathrm{pp}$ on a separate estimation half of the historical holdout. Two prospective comparisons with latest-version observation give $40/40$ positive paired differences. The benefit is conditional on the branch's data. Independent branch orders attenuate the gain over latest-version observation. In a preregistered equal-size data control, $P$ exceeds terminal-only by $4.08/13.51\mathrm{pp}$ when host and branch share a training stream, but trails by $1.86/1.16\mathrm{pp}$ when host and branch train on group-disjoint data. A retained-seed extension separates order and data contrasts: independent ordering reduces the shared-data advantage, while SQuAD retains a positive gap. The tested full-model recipe also favors terminal-only training. These results establish a repeatable shared-stream observation effect, rather than a general advantage of earlier host versions, and show why observation policies must be evaluated jointly with branch data and the deployment host.