scoped-factual-knowledge-lifecycle

OUT derived (depth 5)

Created 2026-08-25T03:11:44+00:00

For single-hop factual (entity-relation-object) knowledge specifically, the full lifecycle—distributed corpus acquisition → local MLP key-value storage → parametric recall → contextual supplementation → rank-one correction—forms a coherent, validated closed loop in which each stage is geometrically well-defined and operationally distinct.

Justifications

SL — The scoped lifecycle requires the local storage principle (acquisition→storage), the contextual edit mechanism (the supplementation step), AND the validated routing (confirming the two channels are genuinely distinct); the gate on multi-hop failure acknowledges that the loop breaks specifically at the compositional/multi-hop boundary rather than at any single factual step.

Antecedents (all must be IN):

  • IN local-storage-distributed-acquisition — Factual knowledge is acquired through distributed corpus exposure (Kandpal's log-linear document-count dependence) but stored in a locally addressable MLP slot (ROME's single-layer FFN edit), revealing a two-phase knowledge pipeline.
  • OUT context-window-ephemeral-edit — The long-context window with near-perfect in-context key-value retrieval functions as an ephemeral, weight-free knowledge editing mechanism: injecting a fact into the 200K-token context is functionally equivalent to a rank-one edit that bypasses the parametric write subspace, providing a complementary write channel with zero persistence cost.
  • OUT knowledge-routing-faithfulness-validated — The parametric/contextual two-channel knowledge architecture is a genuine computational duality rather than a surface-level re-ranking bias, because retrieval context demonstrably inverts the parametric accuracy trend (accuracy increases with document relevance for rare facts) and BM25 recall remains robust independently of parametric scaling.

Unless (any of these IN defeats this justification):