A series of real networks invariants

Mikhail Tuzhilin

arxiv: 2601.02052 · v2 · submitted 2026-01-05 · 💻 cs.SI · math.CO

A series of real networks invariants

Mikhail Tuzhilin This is my paper

Pith reviewed 2026-05-16 18:16 UTC · model grok-4.3

classification 💻 cs.SI math.CO

keywords network centralityLaplacian matrixexponential distributionpower-law distributionreal networksgraph invariantsdegree centralityksi-centrality

0 comments

The pith

Laplacian-based centralities follow exponential distributions in real networks.

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The paper introduces a series of centralities computed using the Laplacian matrix of a network. These generalize the standard degree centrality and the ksi-centrality measure. For real networks, the distributions of these centrality values are exponential, becoming power-law when the index j is set to zero. In contrast, artificial networks produce different distribution shapes. This property offers a potential new way to identify invariants that are characteristic of real-world networks.

Core claim

A parameterized family of centralities is defined from the Laplacian matrix, generalizing degree and ksi-centrality. These centralities display exponential distributions across nodes in real networks, with the special case j=0 yielding a power-law distribution, while artificial networks exhibit distinctly different distributions.

What carries the argument

The central object is the series of Laplacian-based centralities, obtained by considering successive powers or related operations on the graph Laplacian matrix.

If this is right

The new centralities act as additional invariants for real networks.
Exponential distributions provide a statistical marker that separates real networks from artificial constructions.
The j=0 case recovers the power-law behavior known for degree distributions in many real networks.
These measures can be used to analyze and compare the structure of various networks.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

If confirmed across more datasets, these centralities could inspire new network generation models that match the observed distributions.
The approach might extend to other matrix-based measures, such as those from the adjacency matrix, to find similar patterns.
This could help in detecting synthetic or manipulated parts within otherwise real networks by checking distribution fits.

Load-bearing premise

The observed exponential distributions for these centralities are a genuine property of real networks and not due to specific choices in data or parameters.

What would settle it

Finding a substantial collection of real networks where the distributions of these Laplacian centralities deviate from exponential (or power-law for j=0) would falsify the main claim.

read the original abstract

In this article we propose a generalization of two known invariants of real networks: degree and ksi-centrality. More precisely, we found a series of centralities based on Laplacian matrix, that have exponential distributions (power-law for the case $j = 0$) for real networks and different distributions for artificial ones.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Desk Editor's Note private letter to a colleague

The paper generalizes degree and ksi-centrality via a Laplacian-based series and claims exponential distributions hold for real networks but not artificial ones, though the robustness of that split needs more checking.

read the letter

The main point is a proposed series of centralities built from the Laplacian matrix that the author says follow exponential distributions in real networks, with a power-law only when j equals zero, while artificial networks show different patterns. This is framed as a direct generalization of degree and ksi-centrality. The work does a reasonable job laying out the idea that certain matrix-derived measures might serve as invariants separating real from synthetic graphs, which aligns with long-standing questions in network science about what makes real networks special. If the definitions are clean and the pattern survives basic checks, it could give people a new tool for quick characterization. The soft spot is the lack of clear robustness tests. The distinction between real and artificial networks could easily depend on how the artificial graphs are generated or on the specific real datasets chosen. Without comparisons to degree-preserving randomizations or a broader set of null models, it's difficult to know whether the exponential shape is a genuine property or an artifact of the construction. The paper also needs to spell out the exact formulas for each j and show any statistical support for the claimed distributions rather than just stating them. This is the kind of work that might interest researchers already using Laplacian methods or hunting for new centrality invariants. A reader looking for immediate applications or theoretical unification will probably come away wanting more, but the core observation is worth examining. I would send it to peer review so referees can verify the derivations and push on the null-model controls.

Referee Report

2 major / 1 minor

Summary. The manuscript proposes a generalization of degree and ksi-centrality via a series of centralities derived from the Laplacian matrix. It claims these centralities follow exponential distributions in real networks (power-law for the j=0 case) and different distributions in artificial networks.

Significance. If substantiated with explicit definitions, derivations, and tests across diverse datasets and null models, the work could identify useful distributional invariants for distinguishing real networks. The current presentation, however, contains no equations, proofs, datasets, or verification steps, so the potential significance cannot be assessed.

major comments (2)

[Abstract] Abstract: the central claim that the Laplacian-based centralities exhibit exponential distributions specifically for real networks (and distinct ones for artificial networks) is stated without any defining equations for the series, without the Laplacian powering or combination rules for each j, and without reference to any datasets or statistical fitting procedures.
[Abstract] Abstract: no comparison to null models (e.g., degree-preserving randomizations) or robustness checks across network classes is mentioned, leaving open the possibility that the reported distributional distinction is an artifact of the chosen networks or the precise definition of the centrality series.

minor comments (1)

[Abstract] Abstract: the term 'ksi-centrality' appears without definition or citation; a brief explanation or reference would aid readability.

Simulated Author's Rebuttal

2 responses · 0 unresolved

We thank the referee for the detailed comments. We agree that the current presentation lacks necessary mathematical definitions, dataset references, and validation steps, which prevents proper assessment of the claims. We will substantially revise the manuscript to address these issues.

read point-by-point responses

Referee: [Abstract] Abstract: the central claim that the Laplacian-based centralities exhibit exponential distributions specifically for real networks (and distinct ones for artificial networks) is stated without any defining equations for the series, without the Laplacian powering or combination rules for each j, and without reference to any datasets or statistical fitting procedures.

Authors: We acknowledge that the abstract (and manuscript) as currently written does not include the explicit definitions, the Laplacian powering or combination rules for each j, or references to datasets and fitting procedures. In the revised version we will add these elements, including the precise mathematical formulation of the centrality series and the statistical methods used to identify the exponential (or power-law) distributions. revision: yes
Referee: [Abstract] Abstract: no comparison to null models (e.g., degree-preserving randomizations) or robustness checks across network classes is mentioned, leaving open the possibility that the reported distributional distinction is an artifact of the chosen networks or the precise definition of the centrality series.

Authors: We agree that the absence of null-model comparisons and robustness checks is a significant gap. We will incorporate comparisons against degree-preserving randomizations and other standard null models, together with tests across multiple network classes, sizes, and domains, to demonstrate that the reported distributional patterns are not artifacts. revision: yes

Circularity Check

0 steps flagged

No circularity: empirical observation without self-referential derivation

full rationale

The provided abstract and description present the central claim as an empirical discovery: a series of Laplacian-based centralities exhibit exponential (or power-law) distributions specifically on real networks. No equations, parameter-fitting steps, or derivation chain appear in the text. No self-citations, uniqueness theorems, or ansatzes are invoked. The distinction between real and artificial networks is stated as an observed fact rather than a quantity derived from the same data or definitions by construction. This satisfies the default expectation of no significant circularity.

Axiom & Free-Parameter Ledger

0 free parameters · 0 axioms · 0 invented entities

Abstract-only review provides no explicit free parameters, axioms, or invented entities; the central claim rests on unstated assumptions about the Laplacian construction and distribution fitting.

pith-pipeline@v0.9.0 · 5322 in / 1003 out tokens · 22566 ms · 2026-05-16T18:16:54.154639+00:00 · methodology

discussion (0)

Lean theorems connected to this paper

Citations machine-checked in the Pith Canon. Every link opens the source theorem in the public Lean library.

IndisputableMonolith/Foundation/AbsoluteFloorClosure.lean reality_from_one_distinction unclear

?

unclear
Relation between the paper passage and the cited Recognition theorem.

we found a series of centralities based on Laplacian matrix, that have exponential distributions (power-law for the case j = 0) for real networks and different distributions for artificial ones
IndisputableMonolith/Cost/FunctionalEquation.lean washburn_uniqueness_aczel unclear

?

unclear
Relation between the paper passage and the cited Recognition theorem.

ξ^j_i = |out(N_j(i))| / |N_j(i)| (Theorem 1)

What do these tags mean?

matches: The paper's claim is directly supported by a theorem in the formal canon.
supports: The theorem supports part of the paper's argument, but the paper may add assumptions or extra steps.
extends: The paper goes beyond the formal theorem; the theorem is a base layer rather than the whole result.
uses: The paper appears to rely on the theorem as machinery.
contradicts: The paper's claim conflicts with a theorem or certificate in the canon.
unclear: Pith found a possible connection, but the passage is too broad, indirect, or ambiguous to say the theorem truly supports the claim.