{"id":"7e075a01-4ca1-4838-9498-058adc3410e9","arxiv_id":"2501.03474","paper_version":1,"verdict":"UNVERDICTED","confidence":"MODERATE","novelty_score":8.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":0,"one_line_summary":"Benjamini-Schramm limits of Masur-Smillie-Veech random translation surfaces of area g and genus g are Poisson translation planes of intensity 4.","lead":"This research announcement says that random high-genus translation surfaces, sampled with area equal to genus, converge locally to a newly defined 'Poisson translation plane' with intensity 4. It is the flat-surface analogue of known local limit theorems for random hyperbolic surfaces, and proof details are promised in a companion paper.","discovery_kind":"first_principles","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The stated fixed-singularity estimates have the wrong scale: Propositions 1.3 and 1.4 use unit-area radius R√g, which corresponds to area-g radius R g, not the radius R of Theorem 1.1; the local-scale binomial needs R/√g.","rationale":"I read the paper as a research announcement whose central claim is that MSV-random area-g translation surfaces Benjamini-Schramm converge to the Poisson translation plane of intensity 4. For that claim to hold, the fixed-singularity estimates in Section 1.3 must be valid at the unit-area scale R/√g, because the area-g metric is obtained from the unit-area metric by multiplying all distances by √g. Instead, Propositions 1.3 and 1.4 state estimates at the unit-area scale R√g, which is the area-g scale Rg. This is not a subtle gap; it is a direct inconsistency with the scaling normalization the paper itself introduces. The reader's verdict emphasized the deferred random-basepoint and planarity arguments in [BRV]; I agree those are load-bearing, but the scaling error in the stated propositions is more concrete and can be checked immediately. If the intended statement really involves R/√g, then the announcement needs a correction and the subsequent random-basepoint upgrade still has to be supplied. If the intended statement is literally R√g, then the announced estimates do not address the local neighborhoods appearing in Theorem 1.1, and the intensity-4 identification is unsupported by the text. In either case the paper should not be treated as verified; it should be revised or matched against the companion paper [BRV]. I therefore recommend a conditional verdict rather than leaving the issue solely as 'unverifiable from this announcement.'","tokens_in":8011,"tokens_out":26324,"duration_ms":263704,"concrete_test":"Perform the coordinate check from Section 1.2 literally: for X ∼ P_g, push the event {#Σvis(X, σ, R√g) = k} forward to Y = √g · X and record the radius of the image ball; it is Rg, not R. Then recompute the binomial approximation at the local scale L = R/√g: check whether the formula in Proposition 1.4 with R replaced by L√g yields mean 8πR^2 (compatible with Theorem 1.1) or mean 8πL^2/g → 0 (incompatible). The correct local statement should have parameter 4πL^2 = 4πR^2/g; any appearance of 4πL^2/g^2 contradicts the claimed intensity. If changing the scale makes the statement match the theorem, the paper needs only a correction; if not, the announced proof is missing its key estimate.","verdict_should_be":"CONDITIONAL","load_bearing_attack":"The central theorem concerns radius-r neighborhoods in Phyp_{g*}-random surfaces of area g. Section 1.2 defines Y = √g · X for X of area 1, so a ball of Euclidean radius r in Y is the image of a ball of radius r/√g in X. The announced proof states its fixed-singularity estimates at the opposite scale: Proposition 1.4 counts visible singularities in a ball of radius R√g around σ in X, which corresponds to a ball of radius Rg around the corresponding point in the area-g surface. For fixed R this is g times larger than any radius r in Theorem 1.1, so the Poisson limit stated there (mean 8πR^2) is not the limit of the local neighborhoods used for Benjamini-Schramm convergence. The same factor appears in Proposition 1.3, where the event injrad_X(σ) < R√g is the complement of a local injectivity-radius lower bound at scale R/√g. As written, these propositions do not connect to Theorem 1.1. The random-basepoint and planarity arguments in [BRV] are also deferred, but the scaling mismatch is an internal inconsistency independent of that deferral: either the unit-area radius should be R/√g, or the binomial parameter must be rescaled by a factor of g^2. If the latter is what [BRV] proves, the expected local count vanishes and the announced intensity 4 cannot follow.","agreement_with_reader":"partial"},"referee_report":{"model":"deepseek-v4-flash","summary":"The manuscript is a research announcement of a Benjamini-Schramm limit for Masur-Smillie-Veech random translation surfaces with area equal to genus. It claims that for a uniformly random basepoint, the pointed radius-r neighborhood of an area-g surface converges to the radius-r neighborhood of the root in a new object, the 'Poisson translation plane' of intensity 4, constructed recursively from independent Poisson point processes on slit planes. Three fixed-singularity estimates on unit-area surfaces are stated: control of inter-singularity distances (Proposition 1.2), injectivity radius at a singularity (Proposition 1.3), and the distribution of visible singularities (Proposition 1.4, with Corollary 1.5). The paper states that the random-basepoint versions of these estimates and the full proofs are deferred to the companion paper [BRV]. None of the propositions are proved in this manuscript.","tokens_in":8300,"tokens_out":27081,"duration_ms":259483,"significance":"If correct, the theorem would give the first Benjamini-Schramm limit for random translation surfaces under the MSV measure in the area-linear-in-genus regime, and the Poisson translation plane is an explicit and natural limiting object. The intensity-4 prediction is crisp and falsifiable, and the heuristic derivation from total cone angle and planarity is informative. The paper is transparent that the full proofs are in the companion paper [BRV], and it correctly identifies the relevant tools (Aggarwal's volume asymptotics, Siegel-Veech theory, and star surgeries). These strengths make the announced result worth pursuing. However, the stated supporting estimates are not merely unproved; they are stated at a scale that does not match the local scale of Theorem 1.1, and Proposition 1.3 is false as written. The announcement therefore cannot currently be relied upon as a correct summary of the proof.","major_comments":[{"comment":"As stated, Proposition 1.3 is false. In any unit-area translation surface, the metric ball of radius r about any point, if simply connected, contains a Euclidean disk (or a cone disk centered at a singularity) of area at least πr²; since the total surface area is 1, the injectivity radius at any point is at most 1/√π. For fixed R>0 and all g>1/(πR²), the threshold R√g exceeds 1/√π, so the event injrad_X(σ)<R√g has probability 1, contradicting the claimed o_R(1). The intended local statement, if it is what the proof needs, should be at scale R/√g rather than R√g.","section":"§1.3, Proposition 1.3"},{"comment":"The estimates are stated at the wrong scale for Theorem 1.1. A ball of radius r in the area-g surface Y=√g·X corresponds, as the paper notes in Section 1.2, to a ball of radius r/√g in the unit-area surface X. Proposition 1.4 counts visible singularities in a ball of radius R√g in X, which is a ball of radius Rg in the area-g surface; this radius diverges with g, whereas Theorem 1.1 concerns fixed finite radius. If one formally substitutes the local radius r/√g into Proposition 1.4, the binomial probability 4πR²/g becomes 4πr²/g² and the mean is O(r²/g), tending to zero. The stated estimate therefore does not deliver the nonzero local Poisson statistics needed for the Benjamini-Schramm limit. Since the random-basepoint versions are explicitly deferred to [BRV], the announced support for Theorem 1.1 is missing at a load-bearing point.","section":"§1.2–1.3, Propositions 1.3–1.4 and Corollary 1.5"},{"comment":"Theorem 1.1 is a statement about random basepoints, but all three displayed propositions are only for fixed singularities, and the paper explicitly says that the random-basepoint versions appear in the unpublished companion [BRV]; the planarity argument via star surgeries is likewise only sketched. In a research announcement this division of labor can be acceptable, but the manuscript should state prominently that Theorem 1.1 is conditional on [BRV] and should at least formulate the random-basepoint estimates that are actually used, so that the announced theorem can be checked against the companion paper when it appears.","section":"§1.3 and §1.5"}],"minor_comments":[{"comment":"The normalization in equation (1) is inconsistent with the definition in Section 1.1: if the visible singularities from the root form a Poisson process of intensity λ on C, the expected number in a disk of radius r is λπr², not λ·2πr². The heuristic still yields λ=4 if the right-hand side is corrected to λπr² and the per-singularity visible-ball area to 2πr², but the displayed equations should be made consistent.","section":"§1, Eq. (1)"},{"comment":"In the third bullet, 'In [BL21], it shown' should be 'it is shown'.","section":"§1.4, Related literature"},{"comment":"The companion paper is cited inconsistently as both '[BR V]' and '[BRV]'; the citation should be standardized.","section":"References"},{"comment":"The recursive construction of the Poisson translation plane and the description of star surgeries are informal; for an announcement, a precise definition of the intensity (points per unit area in the holonomy plane) and a more detailed statement of the planarity result would reduce ambiguity.","section":"§1.1 and §1.5"},{"comment":"The phrase 'there are (2g−2) singular points on a surface of genus g' should specify that singularities are counted with multiplicity, since the number of singular points depends on the stratum.","section":"§1, heuristic paragraph"}],"recommendation":"major_revision","confidential_remarks":"The scale issue in major comments 1 and 2 is not a matter of taste: Proposition 1.3 is false as stated, and the mismatch between the scale of Propositions 1.3–1.4 and the scale needed for Theorem 1.1 is exactly the kind of error that undermines confidence in the announced proof outline. The authors should correct the statements and make the companion paper available to the referee before a final decision, especially because the main theorem and its proof are entirely deferred to [BRV]. There is also a question of whether a research announcement with no proofs and an unpublished companion paper meets the journal's standards for publication; I leave that to editorial judgment."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Colleague—\n\nYou should know this paper before you cite it. The announced result—a local Benjamini-Schramm limit for MSV-random translation surfaces of area g, with a limiting 'Poisson translation plane' of intensity 4—is exactly the kind of thing the field wants. If true, it's the first local limit theorem for translation surfaces in this regime, and the Poisson plane object is a natural construction. The heuristic intensity computation in Section 1 is coherent, and the paper is candid about being an announcement, with full proofs deferred to [BRV].\n\nThe soft spots, though, are more than missing proofs. The scaling in Section 1.2 is wrong as written. It says a ball of radius R in the area-g surface corresponds to a ball of radius R√g in the unit-area surface. It should be R/√g. That mistake carries through to Propositions 1.3 and 1.4, which state estimates for fixed singularities in a unit-area surface at radius R√g. In the area-g surface, that's radius Rg—g times larger than any fixed radius r in Theorem 1.1. At that scale the claims are not just misplaced; they're false. For example, Proposition 1.3 claims P(injrad_X(σ) < R√g) = o(1), but in a unit-area surface the injectivity radius at any point is bounded by an absolute constant (the systole), so for any fixed R and large g the event has probability 1. Similarly, the binomial in Proposition 1.4 has mean 8πR², but a ball of radius R√g contains ~g² singularities; a Poisson(8πR²) count is off by a factor of g². So either the propositions have a systematic factor-of-g typo and the real estimates are at radius R/√g, or the announced strategy doesn't prove Theorem 1.1. Either way, the announcement as written cannot be checked.\n\nThis is worth saying because the reader's UNVERDICTED is polite: the announcement is not merely unproved, it's internally inconsistent at the level of scaling. The full paper may well fix this, and the underlying approach (Siegel-Veech theory plus star surgeries) is plausible. But I wouldn't cite this version, and I'd be cautious about building on Theorem 1.1 until [BRV] appears.\n\nWho's it for? Teichmüller theorists and anyone working on random surfaces. It deserves a serious referee—the announced result is important enough that an editor should send it out, if only to force a corrected scaling. But the referee's report should ask for the random-basepoint statements and a rescaling of the propositions before the announcement is publishable.\n\nIn short: pass on citing it, but keep an eye on the full paper.","headline":"The announced result is important and new, but the scaling in the announcement is internally inconsistent—the key propositions are at radius R√g in the unit-area surface, which is Rg in the area-g surface, not the fixed local radius in Theorem 1.1.","tokens_in":8852,"tokens_out":12712,"would_cite":false,"duration_ms":113171,"reading_group":"maybe","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":["32G15","30F30","60B10"],"pacs":[],"model":"deepseek-v4-flash","headline":"The paper proves that MSV-distributed random translation surfaces, with area equal to genus, converge locally to a Poisson translation plane with intensity 4.","keywords":["translation surfaces","local weak convergence","MSV measure","Poisson translation plane","large genus asymptotics","saddle connections","flat geometry","random pointed metric spaces"],"falsifier":"If one can compute, for an explicit family of high-genus MSV-random surfaces, the probability that a uniformly random point lies in a ball of radius $R$ containing a closed geodesic of length at most $R$, Theorem 1.1 predicts this probability tends to $0$ for each fixed $R$; a positive $\\liminf$ would refute the claimed planarity. Alternatively, the visible-singularity count from a fixed singularity must have second factorial moment tending to $(8\\pi R^2)^2$, and a different value would rule out the Poisson law.","tokens_in":7677,"feed_emoji":"📐","tokens_out":13394,"duration_ms":125404,"temperature":0.7,"pith_summary":"This paper establishes the local limit of a large random translation surface sampled from the natural measure on surfaces of genus $g$ and area $g$. The claim is that as $g\\to\\infty$, a uniformly random point sees a radius-$r$ neighborhood indistinguishable from the root neighborhood of a random object introduced here, the Poisson translation plane of intensity 4. In that plane there is no local topology: the root is a regular point, all singularities are order-one cone points, and the singularities visible from the root form a Poisson point process in the holonomy plane. The intensity 4 is forced by counting visible area around the $2g-2$ singularities against what a typical point sees. If correct, the result gives a concrete answer to what living at a typical point of a high-genus translation surface is like, and provides a new class of local weak limits analogous to the hyperbolic-plane limit of random hyperbolic surfaces.","feed_headline":"Random high-genus surfaces become a Poisson plane","feed_subtitle":"At a typical point, visible cone points follow a Poisson law with intensity 4, and local topology vanishes.","key_machinery":"Two objects do the work. The first is the Poisson translation plane itself, defined by recursive independent slitting and gluing of copies of $\\mathbb{C}$ along rays to points of a Poisson process of intensity 4. The second is the proof mechanism: three estimates on MSV-random unit-area surfaces -- proximity of singularities, injectivity radius near a singularity, and the visible-singularity count -- combined with a planarity argument. Planarity is achieved by composing star surgeries, deformations that turn a short simple closed geodesic near the basepoint into controlled data such as short saddle connections or pairs of homologous saddle connections; the volume of surfaces carrying such data is then bounded using large-genus asymptotic volume formulas and standard saddle-connection counting estimates.","core_discovery":"The central claim is Theorem 1.1: the pointed MSV-random surfaces $P^{\\mathrm{hyp}}_{g*}$, with genus and area both $g$, converge weakly in the pointed Gromov-Hausdorff-Prokhorov topology to the law of the Poisson translation plane with intensity 4. In this plane the root is regular almost surely, every singularity has cone angle $4\\pi$, and the holonomies of geodesics from the root to visible singularities form a Poisson point process on $\\mathbb{C}$; at each visible singularity, an independent copy of the same process is attached recursively, and these sectors fill the plane. The paper supports this with a fixed-singularity estimate: the number of singularities visible from a fixed singularity within radius $R\\sqrt{g}$ is asymptotically Poisson with mean $8\\pi R^2$.","pith_inferences":["The constant 4 is essentially forced by topology and normalization: the heuristics equate $2g-2$ singularities each contributing $4\\pi r^2$ of visible area with $g$ points each seeing $\\lambda\\cdot 2\\pi r^2$, so any family of flat surfaces whose area grows linearly in genus should exhibit a Poisson translation plane with intensity determined by this same ratio.","Because the random-basepoint versions of the three estimates are deferred to [BR V], the announced theorem currently stands on that companion proof; if the transfer from fixed singularities to a random basepoint fails, the limit could differ even though the individual estimates are true.","A direct computational check on explicit high-genus families, such as square-tiled surfaces, could test the Poisson prediction by comparing empirical visible-singularity counts and holonomy distributions with the intensity-4 model.","If the Poisson translation plane picture is robust, other natural local measures on moduli space may have the same universal local behavior, with the intensity set by their large-genus saddle-connection constants."],"forward_implications":["For each fixed $R>0$, the expected number of singularities visible from a typical point within distance $R$ is asymptotic to $4\\pi R^2$.","Counting visible singularities from a fixed singularity in a ball of radius $R\\sqrt g$ converges in distribution to $\\mathrm{Poisson}(8\\pi R^2)$.","The limiting pointed surface is almost surely planar, so a typical point is not wrapped by short simple closed geodesics.","Rescaling the metric by a factor $s$ changes the limiting intensity to $4/s^2$, giving a family of local limits as area grows linearly with genus.","Quantitative bounds on the probability that a random point has small injectivity radius follow from the same estimates, giving statistical control of local geometry at all scales."],"supporting_citations":[{"why":"Supplies the finite measure on each genus-$g$ moduli space that defines the random model.","marker":"[Mas82]"},{"why":"Provides the companion construction of the same finite measure and its ergodic properties.","marker":"[Vee86]"},{"why":"Defines the notion of local weak convergence of finite random spaces used in Theorem 1.1.","marker":"[BS01]"},{"why":"Metrizes the space of pointed metric measure spaces, fixing the topology in which weak convergence is asserted.","marker":"[Khe20]"},{"why":"Provides the large-genus volume asymptotics for strata that drive the three statistical estimates.","marker":"[Agg20]"},{"why":"Provides the asymptotic constants for counting saddle connections used in the estimates.","marker":"[Agg19]"},{"why":"Handles short simple closed geodesics lying in cylinders, needed for the planarity argument.","marker":"[MRR22]"},{"why":"Contains the random-basepoint versions of the three propositions and the full construction of the Poisson translation plane.","marker":"[BR V]"}],"fun_headline_variants":["Random high-genus surfaces converge to a Poisson plane","As genus grows, random surfaces become Poisson","Poisson plane: limit of random high-genus surfaces","High-genus randomness yields a Poisson plane"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The announced limit rests on transferring estimates proven at a chosen cone point to a genuinely random point, and on the limiting surface being planar; the transfer is deferred to a companion paper.","fun_headline_variants_meta":{"raw":{"variants":["Random high-genus surfaces converge to a Poisson plane","As genus grows, random surfaces become Poisson","Poisson plane: limit of random high-genus surfaces","High-genus randomness yields a Poisson plane"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000979,"raw_usage":{"total_tokens":4120,"prompt_tokens":872,"completion_tokens":3248,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":488,"completion_tokens_details":{"reasoning_tokens":3188}},"tokens_in":488,"tokens_out":3248,"duration_ms":25404,"temperature":1.0,"reasoning_tokens":3188,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-10T21:53:30.192306+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"If one can compute, for an explicit family of high-genus MSV-random surfaces, the probability that a uniformly random point lies in a ball of radius $R$ containing a closed geodesic of length at most $R$, Theorem 1.1 predicts this probability tends to $0$ for each fixed $R$; a positive $\\liminf$ would refute the claimed planarity. Alternatively, the visible-singularity count from a fixed singularity must have second factorial moment tending to $(8\\pi R^2)^2$, and a different value would rule out the Poisson law.","supporting_citations":[],"review_version":1}