Neighbor Expansion
A graph retrieval strategy that, after an initial recall of seed nodes, traverses one or more hops of neighbors along the graph's edges and incorporates the associated nodes into the context. It uses explicit structural relations to supplement semantic similarity retrieval.
Timeline (last 90 days)
Neighbor expansion is a strategy in graph retrieval: after initially recalling seed nodes, it traverses one or more hops of neighbors along the graph's edges, bringing associated nodes into context using explicit structural relations rather than semantic similarity.
In LExGraph, hybrid retrieval is responsible for locating seeds among memory nodes, and neighbor expansion supplements this with structural associations, forming a two-stage evidence acquisition process of "locate first, then diffuse".
All Facts (2)
Neighbor expansion is a strategy in graph retrieval: after initially recalling seed nodes, it traverses one or more hops of neighbors along the graph's edges, bringing associated nodes into context using explicit structural relations rather than semantic similarity.
50%UnverifiedIn LExGraph, hybrid retrieval is responsible for locating seeds among memory nodes, and neighbor expansion supplements this with structural associations, forming a two-stage evidence acquisition process of "locate first, then diffuse".
50%