REVIEW 3 major objections 4 minor 3 references
Analysis of quantities determining the critical inverse temperature in the annealed Potts model with Pareto vertex weights
T0 review · 3 major / 4 minor · reviewed 2026-08-05 · deepseek-v4-flash
Pith's one-line read For the annealed Potts model with Pareto vertex weights, the critical parameter obeys t_c < 2 ln(q−1), sharpened to 2(τ−2)/(τ−1) ln(q−1), with companion bounds t'_c < 3/2 ln(q−1) and t''_c < ln(q−1), all tight as τ→∞.
desk verdict Solid analytic bounds for the annealed Potts critical point with Pareto weights; the proof is clean but leans on imported uniqueness theorems from [1], and the q↓2 asymptotics are heuristic. read the letter →
The pith
A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.
The reading
What carries the argument
The central object is the function K(t) from [1] (defined in (6)), whose unique positive zero is t_c; its derivatives K' and K'' have unique zeros t'_c and t''_c, which are respectively the tangency point of F_0 through the origin and the inflection point of F_0. For Pareto weights, K, K', K'' admit closed representations (18)–(20) involving one common integral D(t) = ∫_1^∞ w^{−τ+1}/(e^{tw}+q−1) dw. The proofs use the sign criteria (14)–(16) — K(t)<0 for t<t_c and K(t)>0 for t>t_c, and similarly for the derivatives — plus a convexity lower bound on D obtained from the convexity of 1/(e^z+1). The sharpened bound for t_c uses the exact identity K(2 ln(q−1))/K'(2 ln(q−1)) = 2 ln(q−1)/(τ−1), yie
What would settle it
Compute K(t) at t = 2 ln(q−1) from the explicit Pareto representation (18)–(21) for a single pair (τ,q), e.g. τ=5, q=10; the claim t_c < 2 ln(q−1) is equivalent to K(2 ln(q−1)) > 0, so a value ≤ 0 for any τ≥4, q>2 would refute the main theorem.
Extended reading notes
Core claim
The paper proves that for all τ≥4 and q>2 the three positive zeros t_c, t'_c=t_b, t''_c=t_* of K, K', K'' obey 0 < t''_c < t'_c < t_c < ∞, with t_c < 2 ln(q−1), t'_c < 3/2 ln(q−1), t''_c < ln(q−1), and the sharper t_c < 2(τ−2)/(τ−1) ln(q−1). The proof uses explicit Pareto representations of K, K', K'' and convexity of 1/(e^z+1); the sharpened bound follows from an exact Newton-step identity at t = 2 ln(q−1). Sharpness is certified by the homogeneous limit τ→∞, where t_c = 2 ln(q−1), t''_c = ln(q−1), t'_c = T(q) solves (27). The paper also derives the large-q limits (31) and the τ-dependent q↓2 asymptotics of t''_c in (35).
Load-bearing premise
The paper assumes from the companion study [1] that for every q>2 and τ≥4 the functions K, K' and K'' each have exactly one positive zero, with t'_c = t_b and t''_c = t_*, and that F_0'' changes sign once; if any of these existence facts fail, the bounds do not apply.
Editorial extensions
If this is right
- The transition parameter t_c is confined to an explicit interval for all τ≥4 and q>2; combined with the conjectured lower bound 2(τ−5)/(τ−4) ln(q−1) (proved for large q), this gives a two-sided estimate whose width shrinks as τ grows.
- Through γ_c = t_c/F_0(t_c) and the sandwich bound for F_0, the log bounds translate into explicit upper bounds on the critical inverse temperature β_c itself, uniformly in q.
- The large-q limits (31) fix the asymptotic shape of the critical curve: t_c ~ 2(τ−2)/(τ−1) ln(q−1), while t'_c and t''_c both approach ln(q−1).
- The q↓2 behaviour of t''_c has four distinct regimes depending on τ (τ=4, 4<τ<5, τ=5, τ>5), with decay from exponentially small to linear in ln(q−1), which must be reproduced by any numerical computation of the phase diagram.
- The bounds are tight in the homogeneous limit, so they cannot be improved without using more information about the weight distribution than the Pareto tail exponent.
Reading between the lines
- The same single-integral-plus-convexity device may extend to other heavy-tailed weight densities; the moment-based numerical observation (86) suggests a general two-sided bound t_c ∈ (2 μ_3/μ_4 ln(q−1), 2 μ_0/μ_1 ln(q−1)) that the paper does not prove.
- The sharp lower-bound conjecture (32) could plausibly be proven by running the same Newton-step argument from a left endpoint, which would turn the conjecture into a theorem and give a uniform relative-width bound for all τ≥4.
- Since t'_c and t''_c are geometric features of F_0 (tangency and inflection), the new bounds constrain the shape of the phase diagram in the companion paper's Figure 1, and may help certify first-order transitions for finite q without solving the full fixed-point equations.
- One can test whether the constant 2(τ−2)/(τ−1) is determined by the tail exponent alone or by further moments, by comparing with the exponential-weight case from [2] and with light-tailed distributions.
Editorial analysis
A structured set of objections, weighed in public.
Referee Report
Summary. The paper analyzes the quantities t_c, t'_c, t''_c that determine the critical inverse temperature of the annealed q-state Potts model on sparse rank-1 random graphs with Pareto vertex weights of exponent τ≥4. Building on the companion paper [1], which establishes existence, uniqueness, and ordering of these zeros, the author derives explicit upper bounds: t_c<2 ln(q−1), t'_c<3/2 ln(q−1), t''_c<ln(q−1), together with the sharpened bound t_c<2(τ−2)/(τ−1) ln(q−1) and an improved bound for t'_c via an auxiliary quantity T. The paper also proves large-q limits for all three quantities, analyzes the homogeneous τ→∞ limit, and presents a leading-order analysis of t''_c as q↓2 with four distinct decay regimes depending on τ.
Significance. If the results are correct, they give simple, parameter-free analytic control of the critical inverse temperature for a nontrivial family of heterogeneous Potts models. The main inequalities are explicit and sharp in the homogeneous limit, and the large-q limits are proved rigorously. The paper also provides explicit asymptotic forms for t'_c in the homogeneous case, including monotonicity of T(q) and T(q)/ln(q−1). These are useful and publishable contributions. The main caveat is that the paper relies substantially on the companion paper [1] for the fundamental uniqueness and sign properties that underpin the proof strategy.
major comments (3)
- [Section 6, Eq. (91)] The central bounds depend on the sign criteria t<t_c ⇔ K(t)<0, t<t'_c ⇔ K'(t)<0, and t<t''_c ⇔ K''(t)<0, plus the ordering 0<t''_c<t'_c<t_c. These properties are not proved in the manuscript; they are imported from [1, Thm. 1.14, Thm. 1.21] and, for K' and K'', only inferred 'from the proof' of [1, Thm. 1.14]. In particular, the Newton sharpening in §5, Eq. (63), uses convexity of K on [t_c,∞), which requires t_c>t''_c and K''>0 there. If the imported uniqueness/sign statements fail, the objects t_c, t'_c, t''_c are not even well-defined and the inequalities (22),(24) do not follow from evaluating K at specific points. Since the paper has the representations (18)–(20) and, for t''_c, the Φ-characterization (33), a self-contained lemma establishing these facts for the Pareto case would remove the load-bearing dependence on [1].
- [Section 7] The displayed inequality γ_c < t_c(1 + 1/(e^{t_c}−1)) does not follow from the lower bound in (90), F_0(t)>1−q/(e^t+q−1). Correctly, 1/F_0(t)<1+q/(e^t−1). The factor q is missing. Consequently, the large-q estimate below (91) should involve a correction O((q−1)^{1−α}) rather than O((q−1)^{−α}) when α is chosen as in (92). The qualitative conclusion of (93) can be recovered by redefining the exponent, but the displayed equations need correction.
- [Section 2] The q↓2 asymptotic regimes for t''_c are derived by leading-order matching of T_1 and T_2 with no error bounds. As written, the four decay behaviours in (35) are not proved to the standard of a theorem; they are heuristic unless residual terms are controlled. The paper does state 'leading-order' in §7, but the abstract and introduction present these as results. Either add error estimates or explicitly label this section as a conjecture/heuristic analysis. The main bounds in §§4–5 do not depend on this section, but it is part of the advertised contribution.
minor comments (4)
- [Abstract] The conjecture in the Introduction/Section 2 is stated for τ≥4, but the expression 2(τ−5)/(τ−4) is undefined at τ=4. Section 6 correctly states the conjecture for τ≥5. Please correct the earlier statement.
- [Abstract] The phrase 'large-q behaviour of t_c, t_c and t''_c' should read 't_c, t'_c and t''_c'. There is also a typo 'characerization' in Section 7.
- [Section 3] The text says the proof of (18) is given in [1], Appendix C. It would be helpful to include a direct derivation for completeness, especially because the sign properties in Section 4 use the detailed form of the coefficient multiplying D.
- [Section 6] The inequality F_0(t)<1 is immediate, but the lower bound in (90) should be stated as strict for t>0; the subsequent use of the bound holds, but the strictness should be explicit.
Circularity Check
No circular derivation; analytic bounds rely on independent uniqueness theorems from prior work.
full rationale
The paper's central results, the bounds (22) and (24), are obtained by analytic manipulation of the explicit representations (18)–(20) in Section 3, which are re-derived from the definition of K via partial integration for the Pareto density. The inequalities are proven by direct convexity estimates (e.g., equations (42)–(61)) and Newton-iteration identities (25)–(26), with no fitted parameters and no data-driven prediction. The only imported content is the existence, uniqueness, and ordering of the zeros t_c, t'_c, t''_c (equations (11), (14)–(16)), taken from cited prior work [1] (Theorems 1.14 and 1.21, Sections 7.1–7.3). This citation is self-citation (the present author is also an author of [1]), but it is not circular: those theorems are parameter-free proofs whose assumptions (q≥3, τ≥4) do not include the target bounds, and they serve only to define the objects being bounded. The paper explicitly labels the q↓2 asymptotics as leading-order approximations, not definitive predictions. Thus no reduction of the claimed results to their inputs by construction exists; the score reflects minor self-citation and a lack of full self-containedness, not actual circularity.
Assumptions & free parameters
assumptions (4)
- domain assumption Positive zeros of K, K', K'' exist and are unique, with t'_c = t_b and t''_c = t_*
- domain assumption For Pareto weights with tau >= 4 and all q > 2, F0'' has exactly one positive zero, and 0 < t''_c < t'_c < t_c < infinity
- domain assumption The annealed Potts model on rank-1 random graphs is represented by the pressure p(t, gamma) and conditions F0(t) = t/gamma and p(t, gamma) = p(0, gamma)
- domain assumption Representation (18) of K in the Pareto case
Cite this review
Pith. "Pith review of Analysis of quantities determining the critical inverse temperature in the annealed Potts model with Pareto vertex weights." pith.science (2026). https://pith.science/paper/LHZ4WWEG
@misc{pith2026250821409,
author = {Pith},
title = {Pith review of: Analysis of quantities determining the critical inverse temperature in the annealed Potts model with Pareto vertex weights},
year = {2026},
howpublished = {\url{https://pith.science/paper/LHZ4WWEG}},
note = {Machine review of arXiv:2508.21409}
}
abstract
We consider in this work the crucial quantity $t_c$ that determines the critical inverse temperature $\beta_c$ in the $q$-state Potts model on sparse rank-1 random graphs where the vertices are equipped with a Pareto weight density $(\tau-1)\,w^{-\tau}\,{\cal X}_{[1,\infty)}(w)$. It is shown in \cite{ref1} that this $t_c$ is the unique positive zero of a function ${\cal K}$ that is obtained by an appropriate combination of the stationarity condition and the criticality condition for the case the external field $B$ equals 0 and that $q\geq3$ and $\tau\geq4$, see \cite{ref1}, Theorem~1.14 and Theorem ~1.21 and their proofs in \cite{ref1}, Section~7.1 and Section~7.3. From the proof of \cite{ref1}, Theorem~1.14, it is seen that ${\cal K}'$ and ${\cal K}''$ also have a unique positive zero, $t_c'$ and $t_c''$, respectively, and $t_c'=t_b$ and $t_c''=t_{\ast}$, where $t_b$ and $t_{\ast}$ are the unique positive zeros of ${\cal F}_0(t)-t\,{\cal F}_0'(t)$ and ${\cal F}_0''(t)$, respectively. Here, ${\cal F}_0(t)=E\,[W(e^{tW}-1)/(E\,[W]\,(e^{tW}+q-1))]$, and $t_c$, $t_b$ and $t_{\ast}$ play a key role in the graphical analysis of \cite{ref1}, Section~5.1 and Figure~1. Furthermore, $\gamma_c=\exp(\beta_c)-1$ and $t_c$ are related according to $\gamma_c=t_c/{\cal F}_0(t_c)$. We analyse $t_c$, $t_c'$ and $t_c''$ for general real $\tau\geq4$ and general real $q>2$ by an appropriate formulation of their defining equations ${\cal K}(t_c)={\cal K}'(t_c')={\cal K}''(t_c'')=0$. Thus we find, along with the inequality $0<t_c''<t_c'<t_c<\infty$, the simple upper bounds $t_c<2\,{\rm ln}(q-1)$, $t_c'<\frac32\,{\rm ln}(q-1)$, $t_c''<{\rm ln}(q-1)$, as well as certain sharpenings of these simple bounds and counterparts about the large-$q$ behaviour of $t_c$, $t_c$ and $t_c''$. We show that these bounds are sharp in the sense that they hold with equality for the limiting homogeneous case $\tau\to\infty$.
Figures
Reference graph
Works this paper leans on
-
[1]
Annealed Potts models on rank-1 inhomogeneous random graphs
C. Giardin` a, C. Giberti, R. van der Hofstad, A.J.E.M. Janssen, and N. Maitra, Annealed Potts models on rank-1 inhomogeneous graphs, arXiv: 2502.10553v1, 14 Feb. 2025
work page Pith review arXiv 2025
-
[2]
A.J.E.M. Janssen, The critical temperature in the annealed Potts model with exponential vertex weights, Eurandom preprint series, 2025-08
work page 2025
-
[3]
Bullen, Handbook of means and their inequalities, Springer, 1987
P.S. Bullen, Handbook of means and their inequalities, Springer, 1987. Appendix A. Proof of (76) We show that forq >2 ψ(q−1)>0> ψ((q−1) 3/2),(A1) where ψ(y) =y(1 + lny) +q−1− 1 q (y+q−1) 2 , y≥1.(A2) As to the first inequality in (A1), we letx=q−1>1, and we compute ψ(q−1) = 2x+xlnx− 4x2 1 +x = x 1 +x ((x+ 1) lnx−2x+ 2).(A3) It is easy to show that (x+ 1) ...
work page 1987
Reviewed August 5, 2026 · model on record in the stance chip above.
Discussion (0). Continue with ORCID to comment.