{"id":"5ecb4c74-37d0-43e1-b1f0-83a4674d8bfd","arxiv_id":"1908.03969","paper_version":1,"verdict":"ACCEPT","confidence":"MODERATE","novelty_score":7.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":0,"one_line_summary":"All Galois quartic fields have shapes that are orthorhombic for V4 groups (with regularized equidistribution) and tetragonal for C4 groups (with explicit per-shape counting asymptotics).","lead":"Galois quartic number fields are shown to have four possible lattice-shape families, depending on their Galois group and on whether the prime 2 ramifies. For Klein-four fields the shapes equidistribute in a regularized sense, while for cyclic-four fields each shape occurs with an explicit X^{1/3} asymptotic.","discovery_kind":"extension","skeptic_critique":{"model":"deepseek-v4-flash","headline":"No significant objection identified: the sieve limit interchange in §5.2 is terse but standard, and the constants in Theorem C check out.","rationale":"I read the full manuscript and focused on the V4 counting proof, which is the load-bearing part for Theorem C. The bijection between V4-quartic fields and strongly carefree triples is carefully set up, and the region R(N,r1,r2) correctly encodes the shape constraints, including the s? factors and the ordering conditions. The density computations in Lemma 5.7 match the known density 1 − 6p^{-2} + 8p^{-3} − 3p^{-4} for strongly carefree triples. The finite-sieve Corollary 5.9 follows from the Principle of Lipschitz and Proposition 5.6, and the constants s?/48 were verified. The passage to the full set of congruence conditions is the most delicate step; I checked that the union-bound argument works, that the bad-set counts are O(N/p^2) up to negligible lower-order terms, and that the limsup/liminf sandwich is valid. I did not find a gap that would change the constants or the equidistribution statement. The C4 half rests on cited parametrizations and integral-basis results, which are published and standard; the reductions in §7 to the arithmetic functions FΣ,U are correct, and the constants in Theorem 7.1 are consistent with Proposition 7.6. The only issue I found is the sign in the printed product in Theorem E in the introduction, which is corrected by the proof in Theorem 7.1. This is a typographical error in a secondary statement and does not affect the central claim. Therefore the reader's ACCEPT verdict stands unchanged.","tokens_in":35133,"tokens_out":56706,"duration_ms":532490,"concrete_test":"Recompute the tame constant in Theorem E from equations (7.14)-(7.16) and Proposition 7.6 for A = 1: the asymptotic should be N^(nr)(X) = (1/4) CΣ X^{1/3} + o(X^{1/3}) with CΣ = L(1,χ5)H(1) > 0. If the printed product ∏_{p≡1(4)} (1 − 2/p)(1 − 1/p) is used instead, the constant vanishes, which would make Theorem E false. The check distinguishes a sign typo from a genuine mathematical error in the constant.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The paper's central claim (Theorem C) rests on the sieve limit interchange in §5.2. I examined equations (5.1)-(5.2) closely. The finite-sieve asymptotic (Corollary 5.9) is correctly derived from the Principle of Lipschitz with the stated densities: δ∞ = 1/2 for the sign condition, δ2 = s?/32 for the case-dependent conditions at 2, and 1 − 6p^{-2} + 8p^{-3} − 3p^{-4} for odd primes. The passage from finite Y to all primes is a standard union-bound argument: for fixed Y, triples in L?(Y) that fail a condition at some p > Y lie in W_p, and the sum of their counts is O(N/Y) up to a lower-order term, which vanishes as N → ∞ and then Y → ∞. The bound #R_{W_p}/N = O(p^{-2}) can be justified by writing the offending coordinates as p times a cofactor and applying the Lipschitz estimate to the scaled region of volume N/p^2 with boundary O((N/p^2)^{2/3}); summed over p > Y this gives O(1/Y + N^{-1/3}Y^{-1/3}), which is negligible in the double limit. The limsup/liminf sandwich (5.1)-(5.2) is therefore valid, and the headline constants 5/48 and 1/6 emerge correctly from the case constants 1/96 and 1/24 (including the factor 4 from the square of the logarithmic ratio). The only internal inconsistency I found is in the introduction's Theorem E, where the Euler product is printed as ∏_p (1 − f_A(p)/p)(1 − 1/p); for primes p ≡ 1 (mod 4), p ∤ A, this factor would be 1 − 2/p, whose product over those primes is zero, contradicting the positive constants proved in Theorem 7.1. The body of the paper uses (1 + f_A(p)/p), and the proof of Theorem 7.1 gives positive asymptotics. This is a typographical slip in a peripheral statement, not a defect in the central equidistribution argument.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper determines the shapes of all Galois quartic number fields, splitting into four families according to Galois group C4 or V4 and tame versus wild ramification. In the V4 case, shapes are shown to form two-dimensional families of orthorhombic lattices, with explicit side ratios in terms of the three quadratic subfield discriminants, and the main Theorem C establishes a regularized equidistribution statement with explicit constants: the number of fields with shape in a compact continuity set is asymptotic to C_wild μ_oC(W) X^{1/2} or C_tame μ_oI(W) X^{1/2}, with C_wild = 5/48 and C_tame = 1/6 times the displayed Euler product over odd primes. In the C4 case, shapes are discrete tetragonal lattices determined by a ramification ratio, and Theorem E/Theorem 7.1 provides asymptotic counts for fields of a given shape. The proofs are built on Williams' integral bases, Conway-Sloane conorm diagrams, a bijection with strongly carefree triples, the Principle of Lipschitz, a sieve following Davenport-Heilbronn, and for the C4 case the Wirsing-Odoni and Wiener-Ikehara Tauberian methods.","tokens_in":35496,"tokens_out":5584,"duration_ms":58449,"significance":"If the results hold, this is a substantial contribution to the study of shape distributions of number fields, extending the cubic and S_n results of Terr and Bhargava-Harron to the Galois quartic setting. The paper contains no fitted parameters: all constants are derived from explicit integral bases, Gram matrices, and sieve densities, and the V4 result is tested against Baily's independent count of V4 fields. The regularized equidistribution result gives a concrete explanation for the appearance of log-squared terms in counting V4 fields, and Theorem A's complete-invariant statements are of independent interest. The analytic number theory is standard but carefully assembled, with the reduction to strongly carefree triples and the case-by-case integral-basis computations forming a solid and reproducible core.","major_comments":[{"comment":"The Euler product defining CΣ_A is printed as ∏_p (1 − f_A(p)/p)(1 − 1/p). For primes p ≡ 1 (mod 4) with p ∤ A this factor equals (1 − 2/p)(1 − 1/p), so the product over these primes diverges to 0; this contradicts the positive constants stated in Theorem 7.1 and equation (7.18), where the factor is (1 + f_A(p)/p)(1 − 1/p). Theorem E should be corrected to match (7.18).","section":"§1.2, Theorem E"}],"minor_comments":[{"comment":"The line 'Let s(ii) = 2 and s(i) = s(ii) = 1' is internally inconsistent; it should read s(i) = s(iii) = 1 and s(ii) = 2, as used in Theorem 5.10 and in the subsequent substitutions for the counting regions.","section":"§5.1"},{"comment":"The name 'Revera-Guaca' is a typo for 'Rivera-Guaca' as in the reference list; please correct it in the remark.","section":"§1.1, Remark 1.1(b)"},{"comment":"The arXiv identifier for [Har19] is printed as 'arXiv:11907.07209'; this should be 'arXiv:1907.07209'.","section":"References, [Har19]"},{"comment":"The term 'subperbase' appears to be a typo for 'superbase': the sentence should say that a quadruple not necessarily satisfying the obtuse condition is called a superbase.","section":"§3.1"}],"recommendation":"minor_revision","confidential_remarks":"The paper is well within scope and the central derivations are sound. The only substantive defect is the sign error in the introduction's statement of Theorem E, which contradicts the correct formula in Theorem 7.1; once that is corrected, I would be happy to see the paper accepted."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Colleague,\n\nThe short version: this paper completes the shape classification for Galois quartic fields and gives the first nontrivial equidistribution result for shapes in degree 4. V4 shapes live in two two-dimensional families (base-centered and body-centered orthorhombic lattices); the authors prove regularized equidistribution with explicit constants 5/48 and 1/6 times an Euler product. C4 shapes are discrete tetragonal lattices, parameterized by a ramification ratio, and the paper gives asymptotic counts for each shape.\n\nWhat's genuinely new: Terr's thesis handled Galois cubics, where all shapes collapse to hexagonal. Here the Galois quartic case has infinite families, and the equidistribution statement (Theorem C) is not in the prior literature. The proof is a clean reduction to counting strongly carefree triples with congruence conditions, using the Principle of Lipschitz and a Davenport–Heilbronn-style sieve. The explicit integral bases and conorm diagrams are worked out carefully; the constants emerge from the case analysis rather than being fitted. The C4 counting via sums of two squares and Wirsing–Odoni is standard but effective, and the arithmetic proportions (depending on p mod 4) are a nice observation that the distribution is not coming from a Lie group action.\n\nThe soft spots are mostly about terseness and imported facts. Section 5.2 jumps through the sieve limit interchange quickly; the stress-test note says the missing step is a standard union bound, and I agree the argument is fine, but a referee will want that written out. The paper also leans on earlier classifications (Williams integral bases, HHR+86 parametrization, Hudson–Williams, Spearman–Williams) without proof; that's normal for these results but means the reader has to trust a chain of external references. Minor blemishes: the introduction's Theorem E has a sign error in the Euler product (1 − f_A(p)/p instead of 1 + f_A(p)/p), which would make the constant zero for infinitely many p; the body uses the correct plus sign, so it's a typo. Also the reference to arXiv:11907.07209 looks like a typo for 1907.07209. None of this affects the main results.\n\nWho is this for: anyone working on shapes of number fields, equidistribution of lattice shapes, or counting number fields by discriminant. The paper deserves a serious referee; the central claims are explicit and checkable. I'd send it out and expect minor revisions.","headline":"Completes the Galois quartic shape classification with a clean equidistribution proof; minor sieve terseness and a typo in the introduction don't affect the main results.","tokens_in":36086,"tokens_out":1868,"would_cite":true,"duration_ms":21309,"reading_group":"yes","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":["11R16","11R45","11E12","11P21"],"pacs":[],"model":"deepseek-v4-flash","headline":"This paper classifies the shapes of all Galois quartic fields and proves that V4 shapes equidistribute in orthorhombic lattice spaces with explicit densities, while C4 shapes form discrete families with counted asymptotics.","keywords":["shape of number field","Galois quartic fields","equidistribution","orthorhombic lattices","tetragonal lattices","carefree tuples","quartic fields","discriminant asymptotics"],"falsifier":"Enumerate all $V_4$-quartic fields with discriminant below a large $X$ (for instance $X = 10^{12}$), count those whose shape falls in a fixed box such as $1 \\le x \\le y \\le 2$, and test whether $N_{\\mathrm{wild}}(X,W)X^{-1/2}$ approaches $C_{\\mathrm{wild}}\\,\\mu_{oC}(W)$; any systematic deviation beyond the claimed $o(1)$ would disprove Theorem C.","tokens_in":34909,"feed_emoji":"📐","tokens_out":14091,"duration_ms":120507,"temperature":0.7,"pith_summary":"The paper determines the shape invariant for every degree-four number field that is Galois, and it does so completely: fields with Galois group $V_4$ have orthorhombic shapes, and fields with group $C_4$ have tetragonal shapes. For $V_4$-quartic fields the shape is the lattice whose side ratios are $\\sqrt{|\\Delta_1|}:\\sqrt{|\\Delta_2|}:\\sqrt{|\\Delta_3|}$, where the $\\Delta_i$ are the discriminants of the three quadratic subfields, and the paper proves that these shapes are equidistributed, in a regularized sense, in the two-dimensional spaces of base-centered and body-centered orthorhombic lattices as the discriminant grows. The limiting densities are explicit Euler products: $C_{\\mathrm{wild}} = \\frac{5}{48}\\prod_{p \\text{ odd}}(1 - 6p^{-2} + 8p^{-3} - 3p^{-4})$ and $C_{\\mathrm{tame}} = \\frac{1}{6}$ times the same product. For $C_4$-quartic fields the shape is a tetragonal lattice determined by a ramification ratio, and the paper proves asymptotic formulas for the number of fields of a given shape. A corollary is that the shape is a complete invariant within the totally real $V_4$-quartic fields and within the tamely ramified $V_4$-quartic fields.","feed_headline":"Shapes of all Galois quartic fields are now classified","feed_subtitle":"V4 shapes equidistribute in orthorhombic lattice spaces with explicit densities; C4 shapes are discrete and counted.","key_machinery":"The argument is carried by the conorm diagram of [CS92], a labeling of the points of the Fano plane that encodes a rank-3 lattice up to isomorphism and exposes the combinatorial type of its Voronoi cell. Combined with the integral-basis data of [Wil70] for biquadratic fields, this identifies the $V_4$ shape as the appropriate orthorhombic lattice. The counting proof bijects $V_4$-quartic fields with $\\ast$-strongly carefree triples, estimates the number of such triples in the relevant region with the lattice-point counting lemma of [Bha05], and passes from a finite sieve to all primes by the method of [DH71], producing the Euler product $\\prod_{p\\text{ odd}}(1-6p^{-2}+8p^{-3}-3p^{-4})$. For $C_4$ fields, the parametrization $K = Q(\\sqrt{A(D+B\\sqrt{D})})$ from [HHR+86], together with the integral-basis results of [HW90] and [SW06], reduces the fixed-shape count to sums-of-two-squares representations, handled by a Dirichlet-series factorization and a Tauberian theorem.","core_discovery":"The central claim is that the shape of a Galois quartic field falls into one of four infinite families and that, in the $V_4$ case, these shapes behave like random points in their natural two-dimensional spaces. Precisely, if $K$ is a $V_4$-quartic field with quadratic subfields $Q(\\sqrt{\\Delta_i})$, then its shape is a base-centered orthorhombic lattice with side ratios $\\sqrt{|\\Delta_1|}:\\sqrt{|\\Delta_2|}:\\sqrt{|\\Delta_3|}$ when the prime $2$ ramifies, and a body-centered orthorhombic lattice with the same ratios when $2$ is unramified. The main theorem states that for compact continuity sets $W$, the counts satisfy $N_{\\mathrm{wild}}(X,W)/X^{1/2} \\to C_{\\mathrm{wild}}\\,\\mu_{oC}(W)$ and $N_{\\mathrm{tame}}(X,W)/X^{1/2} \\to C_{\\mathrm{tame}}\\,\\mu_{oI}(W)$, with the constants given above. For $C_4$-quartic fields, the shape is a primitive tetragonal lattice with side ratio $\\sqrt{2}\\cdot r_{\\mathrm{rat}}^{-1/4}$ if $2$ ramifies and a body-centered tetragonal lattice with side ratio $r_{\\mathrm{rat}}^{-1/4}$ otherwise, where $r_{\\mathrm{rat}} = N/|\\Delta_2|$ is the ramification ratio; these shapes sit as discrete points, and the paper gives $X^{1/3}$-asymptotics for the number of fields with a specified shape.","pith_inferences":["The wild families of $C_3^2$-octic fields, mentioned in the paper only in the tame totally real case, should exhibit the same regularized equidistribution if the same conorm-diagram and sieve machinery is applied, with the density constants depending on the local behavior at the ramified prime.","The ratio $C_{\\mathrm{wild}}/C_{\\mathrm{tame}} = 5/2$ suggests a purely local explanation at the prime $2$; interpreting this as the ratio of admissible congruence patterns at $2$ could predict similar constants for other Galois groups with a single wildly ramified prime.","A numerical test of Theorem C at moderate $X$ is feasible because the field count is $X^{1/2}$ up to logs; computing the shape box count for $X \\sim 10^{12}$ would separate the claimed Euler-product constant from nearby alternatives and provide a concrete check of the sieve step.","The same shape-space viewpoint could resolve the general question of which $S_n$-fields have 'random' shapes versus rigid ones: Galois groups with large automorphism groups impose symmetries that shrink the shape space, and the dichotomy seen here between $V_4$ (equidistributed) and $C_4$ (discrete) should persist for other abelian Galois groups."],"forward_implications":["Within the totally real $V_4$-quartic fields, and within the tamely ramified $V_4$-quartic fields, the shape determines the field uniquely.","For $V_4$ fields with shape in a compact set, the count grows like $X^{1/2}$ times the measure of the shape set; both logarithmic factors in the total count $X^{1/2}\\log^2 X$ come from the unbounded shape parameters.","For $C_4$ fields, the discriminant determines the shape, and the number of fields with a fixed shape and discriminant at most $X$ is asymptotic to a constant times $X^{1/3}$, where the constant depends on the shape through the ramification ratio and the set of primes dividing it.","The proportion of $C_4$ fields with ramification ratio $r_{\\mathrm{rat}}$ versus $p\\, r_{\\mathrm{rat}}$ depends on whether $p \\equiv 1$ or $3 \\pmod 4$, so the distribution of these shapes is not governed by a real Lie group action."],"supporting_citations":[{"why":"Supplies the conorm diagram and Voronoi reduction theory used to identify the rank-3 lattice families for the shapes.","marker":"[CS92]"},{"why":"Provides the discriminants and integral bases of biquadratic fields from which the $V_4$ side ratios are derived.","marker":"[Wil70]"},{"why":"States the lattice-point counting lemma used to estimate the number of strongly carefree triples in the relevant region.","marker":"[Bha05]"},{"why":"Gives the sieve technique that justifies passing from finite to infinite congruence conditions in the $V_4$ count.","marker":"[DH71]"},{"why":"Establishes the $X^{1/2}\\log^2 X$ growth that motivates the regularized equidistribution formulation.","marker":"[Bai80]"},{"why":"Parametrizes cyclic quartic fields as $Q(\\sqrt{A(D+B\\sqrt{D})})$, the starting point of the $C_4$ shape computation.","marker":"[HHR+86]"},{"why":"Supplies integral bases for cyclic quartic fields in the ramified cases used to compute the tetragonal shapes.","marker":"[HW90]"},{"why":"Gives the normal integral basis for the unramified cyclic quartic cases needed for the body-centered tetragonal shapes.","marker":"[SW06]"}],"fun_headline_variants":["Galois quartic shapes: V4 equidistribute, C4 discrete","All Galois quartic field shapes now fully classified","V4 quartic shapes hit all orthorhombic lattices","Counting and equidistribution for Galois quartic shapes"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The load-bearing premise of the $V_4$ counting theorem is that imposing the squarefree and pairwise-coprime conditions at all primes simultaneously changes the count from the finite-sieve count by only $o(N)$; if this error were larger, the explicit constants in the equidistribution theorem would not be the true densities.","fun_headline_variants_meta":{"raw":{"variants":["Galois quartic shapes: V4 equidistribute, C4 discrete","All Galois quartic field shapes now fully classified","V4 quartic shapes hit all orthorhombic lattices","Counting and equidistribution for Galois quartic shapes"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000186,"raw_usage":{"total_tokens":1373,"prompt_tokens":1041,"completion_tokens":332,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":657,"completion_tokens_details":{"reasoning_tokens":261}},"tokens_in":657,"tokens_out":332,"duration_ms":4089,"temperature":1.0,"reasoning_tokens":261,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-14T13:57:09.743840+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Enumerate all $V_4$-quartic fields with discriminant below a large $X$ (for instance $X = 10^{12}$), count those whose shape falls in a fixed box such as $1 \\le x \\le y \\le 2$, and test whether $N_{\\mathrm{wild}}(X,W)X^{-1/2}$ approaches $C_{\\mathrm{wild}}\\,\\mu_{oC}(W)$; any systematic deviation beyond the claimed $o(1)$ would disprove Theorem C.","supporting_citations":[],"review_version":1}