knowledge-lifecycle-closed-loop

OUT derived (depth 3)

Created 2026-08-25T03:05:14+00:00

The full knowledge lifecycle in LLMs—distributed corpus acquisition → local MLP key-value storage → parametric recall → rank-one editing correction—forms a closed loop unified by the covariance geometry, where each stage is addressable in the same second-moment coordinate space.

Justifications

SL — The acquisition→storage duality (depth-1), the geometric unification of analysis+editing (depth-2), and the parametric/contextual split (depth-1) together close the loop; ROME's scope limitation (factual entity-relation only) breaks the loop for logical, spatial, and numerical knowledge.

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 covariance-geometry-unifies-analysis-and-editing — The mathematically principled framework for both interpreting (SAE feature extraction, Park polytope analysis) and modifying (ROME rank-one edits) LLM representations is second-moment covariance geometry applied to the residual stream, since C = KKᵀ whitening defines the canonical coordinate system in which all three operations become linear algebra on the same substrate.
OUT parametric-contextual-knowledge-duality — LLMs maintain two functionally independent knowledge channels: parametric (editable via ROME's rank-one weight update) and contextual (supplied via retrieved passages that invert accuracy trends), which operate on separate mechanisms and can be independently manipulated.

Unless (any of these IN defeats this justification):

IN rome-scope-limitation — ROME and Causal Tracing address factual (entity-relation-object) associations only; logical, spatial, and numerical knowledge are explicitly out of scope.