Revision Collisions in Sequential Model Editing: Family-Aligned Write Paths Reduce Factual Interference

Sequential corrections can remain individually successful while collectively corrupting unrelated facts: after 100 Llama-3.1-8B revisions, a single low-rank write path sustains 95.8% edit success yet adds 32.4 percentage points of net exact-match error on fixed, unedited factual probes. Write-path decomposition exposes the source of this interference by fixing MEMIT targets, edited layers, request order, summed read semantics, and a 2,752,512-parameter online budget while varying writable organization; among partitioned policies, online destination assignment is the experimental variable. RevSep routes temporal, relational, and attribute revisions to family-specialized paths with shared overflow, while random, label-shuffled, route-blind, and shared controls separate semantic placement from generic dispersion. At matched partition geometry, RevSep reduces edit-100 probe error from 30.6 to 27.1 points versus random assignment, a paired mean shift of $-3.5$ points (95% bootstrap interval $[-5.3,-1.7]$), and from 31.3 to 27.1 versus label-shuffled assignment ($-4.2$ [$-5.9,-2.5$]), while retaining 96.2% edit success. Relative to the shared path, the reduction is 5.3 points; the shared-to-random-to-aligned ordering also appears with zero subject–relation–template overlap, on Mistral-7B-v0.3, and in mixed-family histories, whereas route corruption and 91% occupancy progressively compress the gain. Write destination is therefore a controllable source of lifelong-edit retention: matching updates to specialized writable state protects factual behavior beyond equal-budget load dispersion.