Source

García-Pérez, Boguñá and Serrano (2017), LIT-869: the flow ⟨k⟩′ = r^ν⟨k⟩ with ν = 2/(γ − 1) − 1 or 2/β − 1, and the phases it implies, derived from the renormalized S¹ parameters under a power-law approximation.

What is asserted, and by whom

That the average degree of the model obeys this flow, with phase boundaries at γ = 3 and β = 2: small-world networks that densify toward a complete graph below them, and sparsening toward a ring above. The authors derive it; it is a statement about the model.

What the record holds toward it

  • THEORY-190 (Active): the coarse-grained link probability tends to the S¹ form with the stated maps as the supernodes’ separation grows large, with relative error of order β times block width over separation.
  • Against the derivation as given: between neighbouring supernodes the union of member links exceeds the S¹ form, by 32–86% in the worked example with two-node blocks (THEORY-190’s note of 2026-10-20). Most of a node’s links lie there, so the flow inherits the excess.

What would confirm or reject it

An exact ν(γ, β, r) that accounts for the excess and keeps the boundaries confirms it; one that moves them refutes the stated form. QUESTION-203 asks the same thing as a question.