PACE: Pre-Write Exposure in Continual Knowledge Editing

Continual knowledge editing has a blind spot: editors decide where and how to write a new fact but typically do not estimate which prior mappings that writing may expose. We formulate this problem as pre-write exposure and study sparse edit-conditioned feed-forward-network support as a prospective exposure signal. This leads to PACE, a bounded-state editor that extracts one activation path before each write and reuses it to coordinate localization, risk-aware replay, sparse solving, and preservation. Across GPT-J, LLaMA-3, and routed Qwen3, support overlap predicts interference more strongly than prompt semantics, while controlled interventions that change overlap under closely matched realized write strength produce corresponding changes in prior-edit damage. Under a fixed replay budget, better exposure ranking is strongly associated with lower forgetting, and the same ranking signal transfers to other editors. At reliability matched to UltraEdit, PACE improves locality by $1.9$–$2.0$ points and reduces forgetting by $1.2$–$1.7$ points on GPT-J and LLaMA-3. These results suggest a broader principle for continual editing: model which prior mappings a write may expose before deciding how to apply it.