{"id":"67a5c8da-4a4e-4edf-a490-9ae9f95e05b2","arxiv_id":"1908.08880","paper_version":2,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":0,"one_line_summary":"For spatial exponential random graphs on Z^2, if the edge-length penalty is strong enough the finite-box Gibbs measures converge to a unique infinite-volume measure, which is exponentially mixing and satisfies a CLT, with a perfect simulation algorithm.","lead":"This paper constructs an infinite-lattice version of a random graph model in which short edges are favored and the local graph shape is penalized, and proves that the infinite model is well-defined and unique when the temperature is low enough. It also provides guarantees for fast spatial mixing, a central limit theorem, and an exact sampling algorithm, which matter for fitting spatial network models.","discovery_kind":"extension","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Uniqueness of the infinite-volume measure is asserted, not proved: Theorem 5.1(2) only constructs one invariant measure, and the perfect-simulation algorithm samples that one.","rationale":"I read the paper in good faith. The existence construction via the clan of ancestors is plausible, and the branching-process domination underlying Proposition 7.3 is standard, though terse; I do not contest the no-percolation conclusion itself. However, the reader's weakest_assumption—finiteness of the clan of ancestors—is necessary for existence, but it is not sufficient for uniqueness. The more exposed part of the central claim is the uniqueness of the infinite-volume measure: Theorem 3.1(1) purports to prove it in a few sentences, and those sentences cite Theorem 5.1(2), which only produces one invariant measure, and a simulation algorithm, which only samples from that measure. Neither rules out other invariant measures. This is a genuine gap in the written argument, independent of whether the percolation estimates are correct. I therefore partially agree with the reader: the clan-finiteness proof deserves scrutiny, but the uniqueness assertion is the single most load-bearing gap. The paper remains plausible and the reader's CONDITIONAL verdict is appropriate; the condition should explicitly include supplying a proof or precise citation for uniqueness of the infinite-volume invariant measure.","tokens_in":18924,"tokens_out":24483,"duration_ms":267327,"concrete_test":"Check the proof of Theorem 3.1(1) against Ferrari et al. (2002): locate the exact theorem, if any, that states that absence of backward percolation implies uniqueness of the invariant measure for the infinite-volume generator (2.6). If no such theorem is cited, attempt to re-derive uniqueness by coupling two copies of η_t started from two different finite-degree initial configurations on the same graphical representation; uniqueness holds iff the two marginal laws converge to the same distribution as t → ∞. If this coupling argument cannot be completed without additional assumptions not stated in the paper, the uniqueness part of Theorem 3.1 remains unproven.","verdict_should_be":"UNCHANGED","load_bearing_attack":"Section 7.4's proof of Theorem 3.1(1) says uniqueness 'is guaranteed by Theorem 5.1-(2) and the construction of the perfect simulation algorithm.' This is not a valid inference. Theorem 5.1(2) constructs one stationary process from the clan-of-ancestors partition and concludes that its marginal is an invariant measure; it says nothing about other invariant measures for the infinite-volume generator A. The perfect-simulation algorithm in Section 6 samples exactly the measure constructed from the clan of ancestors, so it cannot rule out a second invariant measure, such as a limit obtained with different boundary conditions or a different ergodic component. For finite V, irreducibility of the finite-state process gives uniqueness of µ_V, and that argument is fine; but for V = Z^2 no irreducibility, coupling, or Dobrushin-type contraction argument is supplied. Theorem 3.5's exponential mixing is a statement about the constructed finite-volume measures and does not imply uniqueness of infinite-volume invariant measures. Thus the central 'unique infinite-volume measure' claim in the abstract and Theorem 3.1(1) is unsupported as written. The gap is logically independent of whether Proposition 7.3's percolation bound is correct: even granting no backward percolation, uniqueness does not follow from the statements cited in the proof.","agreement_with_reader":"partial"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper defines a family of spatial Gibbs random graphs on finite subsets V of Z^2 with Hamiltonian H_V(x) = sum L(i,j)x_ij + F_V(x), where F_V is a general sufficient statistic satisfying assumptions (A1)-(A2). It proposes a birth-death graphical construction based on marked Poisson processes and the clan of ancestors, and uses this construction to claim an infinite-volume limit measure mu on graphs with vertex set Z^2. The main results are: for beta > beta*, with beta* defined by alpha(beta*) <= 1 and alpha(beta)=8 exp(-beta(M+1))/(1-exp(-beta))^2, there is a unique process with generator A, a unique invariant measure mu, weak convergence of the finite-volume measures mu_V to mu, finite vertex degree almost surely, exponential space convergence, exponential mixing, and a central limit theorem. The paper also proposes a perfect simulation algorithm sampling a finite window of mu. The proofs rely heavily on the earlier papers Fernandez-Ferrari-Garcia (2001) and Ferrari-Fernandez-Garcia (2002).","tokens_in":19208,"tokens_out":13372,"duration_ms":140879,"significance":"If the main claims are correct, this is a useful contribution: it provides a rigorous infinite-volume construction for a spatial exponential random graph model, with an explicit threshold beta* derived from branching-process bounds rather than fitted, quantitative decay estimates, a CLT, and a perfect simulation algorithm. The central derivation is not circular in the sense of fitting parameters: beta* is derived from the branching-process domination and all bounds are stated explicitly. The main weakness is that the uniqueness part of Theorem 3.1 is not actually proved, and several load-bearing technical lemmas are imported from earlier papers without full proofs or precise theorem references. The paper also offers a concrete algorithm, though its presentation contains at least one questionable formula. Overall the manuscript is promising but needs substantial revision before the central uniqueness claim can be accepted.","major_comments":[{"comment":"The proof asserts that uniqueness of mu is 'guaranteed by Theorem 5.1-(2) and the construction of the perfect simulation algorithm.' This inference is not valid. Theorem 5.1(2) constructs one stationary process from the clan-of-ancestors partition and shows that its marginal distribution is invariant; it says nothing about whether there are other invariant measures for the generator A. The perfect simulation algorithm samples exactly the measure constructed from the clan of ancestors, so it cannot rule out a second invariant measure, for example one obtained with different boundary conditions or a different ergodic component. For finite V, irreducibility of the finite-state process gives uniqueness of mu_V, but no such argument is supplied for V=Z^2. Since uniqueness is a central claim in the abstract and in Theorem 3.1(1), this is a load-bearing gap that must be repaired, either by a coupling or Dobrushin-type contraction argument or by weakening the statement to existence of an invariant measure.","section":"§7.4, proof of Theorem 3.1(1)"},{"comment":"The construction of the infinite-volume process is only sketched. The proof says that R[0,t] union R(x) can be partitioned 'following the same procedure as Section 4.2,' and that the generator calculation is 'very similar to Theorem 3.1 in Ferrari et al. (2002).' For V=Z^2 one must justify the Markov property and the generator A^V for an infinite edge set, and one must prove that the constructed process is the unique process with the claimed generator, as Theorem 3.1(1) asserts. The compactness argument cited from Liggett proves existence of an invariant measure, not uniqueness of the process. The authors should either provide the missing argument or state precisely which theorem in the cited papers covers the present setting.","section":"§5 and §7.1, Theorem 5.1"},{"comment":"Proposition 7.3 is the quantitative basis for the threshold beta*, for Theorem 5.1, and for all later estimates in Theorems 3.2, 3.5, and 3.7, but its proof is not self-contained. The construction of the dominating branching process, the claim that the offspring counts are Poisson with mean m({i,j},{k,l}) = 1_{{i,j}~{k,l}} exp(-beta L(k,l) - beta M), and the inequalities in parts (2)-(4) are either asserted or delegated to Ferrari et al. (2002) and Fernandez et al. (2001) without theorem numbers. In particular, the Borel-Cantelli step in (7.4) requires a bound on P( sum_{k,l} b^n_{ij}(k,l) != 0 ), and the domination by the branching process must be established exactly as stated. If the domination is not exactly as claimed, Theorems 5.1 and 3.1-3.7 lose their foundation. The authors should prove Proposition 7.3 in full or provide exact references that cover the present model.","section":"§7.3, Proposition 7.3"}],"minor_comments":[{"comment":"The formula P(tau({i,j}) > t) = 1 - exp(-nu_{ij}(t)) appears to have the survival function reversed; for a nonnegative waiting time one expects P(tau > t) = exp(-nu_{ij}(t)). Please correct this and check the resulting simulation step.","section":"§6, Algorithm 1"},{"comment":"The clause 'independent of hat(A)(Supp_v(f))' should read 'independent of A(Supp_v(f))'; as written, the independence statement is confusing.","section":"§7.2, Lemma 7.2"},{"comment":"In the displayed update for a death time, 'eta_{r_k-1}^{V,x}(l,m)' should be 'eta_{r_k-1}^{V,x}(m,n)'.","section":"§4.2, death step"},{"comment":"Lemmas 7.1, 7.2, and Theorem 6.1 are asserted to follow from specific results in Fernandez et al. (2001) and Ferrari et al. (2002), but no theorem numbers or page references are given. Please add precise references or include the short proofs so that the adaptation to this model can be verified.","section":"§7.2 and §6.1"},{"comment":"The phrase 'finite (infinite) subset' and the surrounding text 'states the mixing property for the finite measure' are inconsistent; please clarify that the bound is stated for the finite-volume measures and also for the infinite-volume measure mu when V is infinite.","section":"§3, Theorem 3.5"}],"recommendation":"major_revision","confidential_remarks":"The manuscript relies very heavily on the authors' own earlier works for the graphical construction and key lemmas. This is not by itself a problem, but the referee should insist that the missing proofs be either included or referenced with precise theorem statements, because the present text does not allow a reader to verify the infinite-volume construction. The uniqueness gap in Theorem 3.1 is the most serious issue; if it cannot be repaired, the theorem should be weakened to existence of an invariant measure and uniqueness of the limit along finite volumes."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"The paper does real work: it takes Ferrari et al.'s spatial Gibbs random graph and extends it to a general class of local sufficient statistics on Z^2, with an explicit beta* from a self-contained branching process bound (Lemma 7.4). The exponential convergence, mixing, and CLT theorems are genuine additions, and the perfect simulation algorithm is concrete. This is more than a cosmetic extension of the 2010 paper.\n\nThe engine is the control of the clan of ancestors in Proposition 7.3. The bounds on space diameter and time length look coherent and the proof, while condensed, gives the essential calculation. Theorems 3.2, 3.5, and 3.7 follow from those bounds in a way that is mostly convincing, provided the omitted lemmas (7.1 and 7.2) hold in this setting; those are cited to earlier work, not proved, and a referee would want enough detail to check them for these Hamiltonians.\n\nThe soft spot is load-bearing: uniqueness of the infinite-volume measure is asserted, not proved. The proof of Theorem 3.1(1) says uniqueness follows from Theorem 5.1(2) and the perfect simulation algorithm. That is not a valid inference. Theorem 5.1(2) constructs one stationary process and its marginal is an invariant measure; the perfect simulation algorithm samples that same marginal. Nothing in those statements rules out other invariant measures for the generator A. For finite V, irreducibility gives uniqueness, and the paper says that; for Z^2, no coupling, Dobrushin-type contraction, or boundary-condition argument is supplied. The gap is independent of whether Proposition 7.3 is correct: even granting no backward percolation, uniqueness of invariant measures does not follow.\n\nThe fix is straightforward in spirit: either prove uniqueness properly, or restate Theorem 3.1(1) as uniqueness of the limit of the finite-volume measures µ_V. The weak convergence and exponential-space-convergence results already give that, so the main applications—mixing, CLT, perfect simulation for the constructed µ—do not collapse. But the abstract and the theorem currently claim more than is shown.\n\nMinor things: the abstract says Z^d while the whole paper is on Z^2; there are a few notational slips in the definition of Supp_e; and the proof of Theorem 6.1 is omitted entirely with a citation. Those are easy to fix.\n\nMy take: this deserves serious refereeing. The constructive machinery is useful, the explicit constants are a real contribution, and the uniqueness gap is repairable by an honest rewording or a real argument. I'd send it to review, with the uniqueness point as the main demand.","headline":"Real technical value in the clan-of-ancestors construction and mixing/CLT bounds, but the claimed uniqueness of the infinite-volume measure is not proved as written.","tokens_in":19699,"tokens_out":3405,"would_cite":true,"duration_ms":36467,"reading_group":"yes","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":["60K35","05C80"],"pacs":[],"model":"deepseek-v4-flash","headline":"The paper proves that a spatial exponential random graph model on the square lattice has a unique infinite-volume Gibbs measure above an explicit inverse-temperature threshold, with finite degrees, exponential mixing, and a central limit…","keywords":["spatial Gibbs random graphs","exponential random graph models","infinite-volume Gibbs measure","graphical construction","clan of ancestors","perfect simulation","exponential mixing","central limit theorem"],"falsifier":"Simulate the backward ancestor process from one edge at $\\beta$ just above $\\beta^*$ and count the number of ancestor generations; the paper's bound says the $n$-th generation average size is at most $\\alpha(\\beta)^n$, so finding even one run with an infinite chain, or an empirical mean that fails to decay geometrically with ratio $\\alpha(\\beta)$, would falsify the no-percolation step on which Theorem 3.1 depends.","tokens_in":18749,"feed_emoji":"🕸️","tokens_out":11489,"duration_ms":104353,"temperature":0.7,"pith_summary":"The paper proves that a spatial exponential random graph model—a Gibbs distribution on graphs embedded in the square lattice that penalizes long edges and graph features such as stars or triangles—has a well-defined infinite-volume limit. Above an explicit inverse-temperature threshold $\\beta^*$, the finite-box measures converge to a unique measure on graphs with vertex set $\\mathbb{Z}^2$, and this measure assigns finite degree to every vertex with probability one. The result is not automatic, because local edge decisions can propagate over unbounded distances through shared vertices; controlling that propagation is the core of the proof. Once the infinite measure exists, the same machinery yields exponential convergence of finite-window expectations, exponential decay of correlations, and a Gaussian central limit theorem for spatial averages of local graph statistics. The accompanying perfect simulation algorithm means the infinite-volume law can be sampled exactly on finite windows, making the model usable in applications.","feed_headline":"Spatial Gibbs random graphs now have a unique infinite-volume law","feed_subtitle":"A graphical construction shows the full-plane measure exists, mixes exponentially, and can be sampled exactly","key_machinery":"The clan of ancestors of an edge is the set of earlier rectangle events in the independent multigraph process that overlap it in time and share a vertex, recursively through all generations; it is the object that determines whether an edge depends on infinitely many past events. The paper dominates this backward percolation process by a multitype branching process whose mean offspring matrix is $m(\\{i,j\\},\\{k,l\\}) = \\mathbf{1}_{\\{i,j\\}\\sim\\{k,l\\}} e^{-\\beta L(k,l)-\\beta M}$, and shows that the total mass of the $n$-th power of this matrix is at most $\\alpha(\\beta)^n$. Finiteness of every clan therefore follows from subcriticality $\\alpha(\\beta) \\le 1$, which is equivalent to $\\beta > \\beta^*$. The clan is used both as the proof vehicle for existence, uniqueness, convergence, and mixing, and as the basis of the perfect simulation algorithm: to sample a finite window, one builds the clan backward in time and then cleans it forward using the acceptance probabilities $Q(\\{i,j\\}|x)$.","core_discovery":"The central discovery is that the spatial Gibbs random graph measure admits a unique infinite-volume extension when $\\beta > \\beta^*$, where $\\beta^*$ is the smallest $\\beta$ such that $\\alpha(\\beta) = 8 e^{-\\beta(M+1)}/(1-e^{-\\beta})^2 \\le 1$. At these temperatures, the Markov birth-and-death process on graphs with generator (2.6) has a unique invariant measure $\\mu$ on graphs over $\\mathbb{Z}^2$; $\\mu$ is the weak limit of the finite-volume measures $\\mu_V$ as $V$ increases to $\\mathbb{Z}^2$, and $\\mu$ is supported on graphs in which every vertex has finite degree. The proof identifies this regime with the absence of backward oriented percolation in a graphical construction: each possible edge birth is a marked Poisson 'rectangle' in space-time, and the dependent process is obtained by cleaning the rectangles that survive the birth-and-death dynamics. Subcriticality of a dominating multitype branching process is exactly the condition $\\alpha(\\beta) \\le 1$. From the finiteness of the resulting clan of ancestors, the paper derives exponential convergence of finite-volume expectations (Theorem 3.2), exponential mixing of the infinite-volume measure (Theorem 3.5), and a central limit theorem for local functions with finite support (Theorem 3.7).","pith_inferences":["The threshold proved here is a sufficient-condition bound; if the domination in Proposition 7.3 is not sharp, uniqueness could persist below $\\beta^*$, and a direct simulation of the ancestor tree would show how much slack exists.","The mixing and CLT statements are written for functions with finite support, so the CLT as stated does not directly cover unbounded statistics such as the total degree in a growing window; extending it requires a separate truncation argument.","Because the constant 8 in $\\alpha(\\beta)$ comes from the square-lattice geometry, the same graphical construction should port to other periodic lattices by changing that constant, with the structural theorems otherwise unchanged.","The perfect sampler gives an unbiased computational null model for spatially embedded networks: sample edge configurations from finite windows and compare observed local statistics against the CLT-calibrated sampling distribution."],"forward_implications":["Finite-window expectations approximate infinite-volume expectations exponentially fast in the distance to the window boundary, so box simulations inherit rigorous error bounds (Theorem 3.2).","Local graph statistics decorrelate exponentially with spatial separation, making the infinite-volume measure strongly mixing (Theorem 3.5).","Spatial averages of bounded local functions obey a Gaussian central limit theorem, so parameter estimation and goodness-of-fit tests can use normal approximations (Theorem 3.7).","The perfect simulation algorithm samples exactly from $\\mu$ on any finite window, with no monotonicity assumption on the edge dynamics.","Almost surely every vertex has finite degree, so the infinite graph is locally finite despite the infinite vertex set."],"supporting_citations":[{"why":"Introduces the clan-of-ancestors graphical representation and the branching-process domination that the paper adapts to spatial Gibbs random graphs.","marker":"Fernandez et al. (2001)"},{"why":"Supplies the perfect simulation scheme via backward construction of the clan and the cleaning procedure used in Algorithms 1 and 2.","marker":"Ferrari et al. (2002)"},{"why":"Defines the original spatial Gibbs random graph model that is a special case of the Hamiltonian considered here and whose existence result this paper extends.","marker":"Ferrari et al. (2010)"},{"why":"Treats the one-dimensional spatial Gibbs random graph that motivates penalizing long edges and provides the context for the $\\mathbb{Z}^2$ extension.","marker":"Mourrat et al. (2018)"},{"why":"Provides the central limit theorem for stationary mixing random fields that Theorem 3.7 uses, once exponential mixing is established.","marker":"Bolthausen (1982)"},{"why":"Gives the compactness argument that produces an invariant measure for the infinite-volume Markov process.","marker":"Liggett (1985)"}],"fun_headline_variants":["Unique infinite-volume law for spatial Gibbs random graphs","Gibbs random graphs: existence, uniqueness, exact sampling","Exponential mixing and CLT for spatial Gibbs graph measure","Spatial Gibbs graphs: a unique law via graphical construction","Perfect simulation for infinite-volume Gibbs random graphs"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"Everything rests on the claim that above the threshold $\\beta^*$ each edge depends on only finitely many earlier random 'ancestor' events, a domination argument whose detailed proof is delegated to earlier papers; if that finiteness fails, the infinite-volume measure is not constructed and the later theorems collapse.","fun_headline_variants_meta":{"raw":{"variants":["Unique infinite-volume law for spatial Gibbs random graphs","Gibbs random graphs: existence, uniqueness, exact sampling","Exponential mixing and CLT for spatial Gibbs graph measure","Spatial Gibbs graphs: a unique law via graphical construction","Perfect simulation for infinite-volume Gibbs random graphs"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000151,"raw_usage":{"total_tokens":1199,"prompt_tokens":943,"completion_tokens":256,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":559,"completion_tokens_details":{"reasoning_tokens":179}},"tokens_in":559,"tokens_out":256,"duration_ms":3326,"temperature":1.0,"reasoning_tokens":179,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-14T11:27:49.618894+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Simulate the backward ancestor process from one edge at $\\beta$ just above $\\beta^*$ and count the number of ancestor generations; the paper's bound says the $n$-th generation average size is at most $\\alpha(\\beta)^n$, so finding even one run with an infinite chain, or an empirical mean that fails to decay geometrically with ratio $\\alpha(\\beta)$, would falsify the no-percolation step on which Theorem 3.1 depends.","supporting_citations":[{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Supplies the perfect simulation scheme via backward construction of the clan and the cleaning procedure used in Algorithms 1 and 2."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Defines the original spatial Gibbs random graph model that is a special case of the Hamiltonian considered here and whose existence result this paper extends."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Treats the one-dimensional spatial Gibbs random graph that motivates penalizing long edges and provides the context for the $\\mathbb{Z}^2$ extension."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Provides the central limit theorem for stationary mixing random fields that Theorem 3.7 uses, once exponential mixing is established."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Gives the compactness argument that produces an invariant measure for the infinite-volume Markov process."}],"review_version":1}