pith. machine review for the scientific record. sign in

arxiv: 2605.14226 · v1 · submitted 2026-05-14 · 🧮 math.NT

Recognition: 2 theorem links

· Lean Theorem

Kodaira-Neron statistics for rational elliptic curves with j-invariant 0 and 1728

Authors on Pith no claims yet
Pith Number pith:YHVAR3RE state: computed view record JSON
4 claims · 10 references · 2 theorem links. This is the computed registry record for this paper; it is not author-attested yet.

Pith reviewed 2026-05-15 02:45 UTC · model grok-4.3

classification 🧮 math.NT MSC 11G05
keywords elliptic curvesj-invariantKodaira-Néron typesadditive reductiontorsion subgroupsisogeny graphsheight asymptoticsrational points
0
0 comments X

The pith

Elliptic curves with j-invariant 0 or 1728 admit explicit asymptotic counts of their Kodaira-Néron reduction types at 3 and 2 when ordered by height.

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

This paper counts rational elliptic curves with j-invariant fixed at 0 or 1728 by their height and computes the asymptotic frequency of each possible Kodaira-Néron type at the prime where bad reduction is forced. For j-invariant 0 these curves always have bad reduction at 3, and the type varies among the additive possibilities; the same holds for j-invariant 1728 at 2. The authors also refine the counts by fixing the torsion subgroup or the full isogeny-torsion graph. These families are special because every bad prime gives additive reduction, so the local pictures are more rigid than for generic curves. A reader would care because such counts give precise information on how often each reduction type occurs in a thin but infinite family of curves.

Core claim

Elliptic curves over Q with j-invariant 0 or 1728 have additive reduction at all primes of bad reduction. In addition, all elliptic curves with j-invariant 0 have bad reduction at p=3 and all elliptic curves with j-invariant 1728 have bad reduction at p=2. We count these curves by height and determine asymptotics for the various Kodaira-Néron types at 3 and 2, respectively. We also give related statistics by holding the torsion subgroup and isogeny-torsion graph constant.

What carries the argument

The explicit Weierstrass models y^2 = x^3 + k for j=0 and y^2 = x^3 + kx for j=1728, ordered by naive height, together with the Kodaira-Néron classification of the singular fiber at the forced bad prime.

If this is right

  • Asymptotic formulas exist for the number of curves of each Kodaira-Néron type at the forced bad prime.
  • The leading constants in these formulas are positive for some types and zero for others.
  • Fixing the torsion subgroup produces a different but still asymptotic distribution of types.
  • Partitioning by isogeny-torsion graph yields separate asymptotic counts within each class.

Where Pith is reading between the lines

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

  • The same counting methods could be applied to other fixed j-invariants that force additive reduction at a small prime.
  • Numerical checks for moderate height bounds could be used to estimate the implied constants without solving the full analytic problem.
  • These families provide a controlled setting in which to test whether reduction type correlates with rank or other global invariants.
  • The results supply a benchmark for heuristics that predict average reduction behavior across all elliptic curves.

Load-bearing premise

The elliptic curves with these fixed j-invariants are sufficiently dense and uniformly distributed by height that their Kodaira-Néron types admit asymptotic counts derived from local conditions at the special prime.

What would settle it

Direct enumeration of all minimal models with j=0 and height below a large explicit bound, followed by computation of the proportion of each Kodaira type at 3; a statistically significant deviation from the predicted leading term would falsify the asymptotics.

read the original abstract

Elliptic curves over $\Q$ with $j$-invariant 0 or 1728 have additive reduction at all primes of bad reduction. In addition, all elliptic curves with $j$-invariant 0 have bad reduction at $p=3$ and all elliptic curves with $j$-invariant 1728 have bad reduction at $p=2$. In this paper we count elliptic curves with $j$-invariant 0 and 1728 by height and determine asymptotics for the various Kodaira-N\'eron types at 3 and 2, respectively. We also give related statistics by holding the torsion subgoup and isogeny-torsion graph constant.

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.

Referee Report

3 major / 2 minor

Summary. The manuscript counts rational elliptic curves with j-invariant 0 (Weierstrass form y² = x³ + B) and j-invariant 1728 (y² = x³ + A x) ordered by naive height on minimal models. It derives asymptotic formulas for the proportions of Kodaira-Néron reduction types at the forced bad prime (p=3 for j=0, p=2 for j=1728) and supplies conditional versions of these statistics when the torsion subgroup or the full isogeny-torsion graph is held fixed.

Significance. If the asymptotics are rigorously established, the work supplies the first explicit leading constants for Kodaira-Néron type distributions inside these two thin CM families. The results are of interest because they must incorporate the constraints of minimality at the fixed prime together with the scaling action on the single parameter (B or A), which can alter naive densities; successful handling of these constraints would be a useful contribution to arithmetic statistics on special families.

major comments (3)
  1. [§3] §3 (height and enumeration): the paper orders curves by the naive height H(E) = max(|c₄|, |Δ|^{1/12}) on minimal Weierstrass models. For the j=0 family the scaling B ↦ u⁶B changes the model while preserving j; the minimality condition at p=3 (v₃(Δ) < 12 after admissible change of variables) restricts the admissible u. It is not shown that the induced distribution on v₃(B) for minimal models still admits the claimed natural density when ordered by H(E); the leading constant in the main asymptotic may therefore be incorrect.
  2. [Theorem 4.2] Theorem 4.2 (asymptotics for Kodaira symbols at p=3): the claimed proportions (e.g., 1/2 for I₀* and 1/6 for II) are derived from the p-adic valuations of B without an explicit error term or Tauberian argument that accounts for the global height bound. Because the family is thin, the usual sieve or lattice-point counting must be justified; the manuscript provides only a sketch and no verification that the error is o of the main term.
  3. [§5.1] §5.1 (conditional statistics): when the torsion subgroup is fixed, the same height ordering is used, yet the torsion condition imposes congruence conditions on B (or A) that may correlate with v₃(B) (or v₂(A)). The paper does not prove that these correlations are negligible in the asymptotic; a separate major-term calculation or explicit example for the 3-torsion case would be needed.
minor comments (2)
  1. [Table 1] Table 1: the column headings for Kodaira symbols should include the corresponding Néron symbol (e.g., I₀*) for clarity; the current notation mixes the two conventions.
  2. [p. 7] p. 7, line 14: the reference to “the standard height” is ambiguous; it should cite the precise definition used in §3 (naive height on minimal models versus the usual H(E)).

Simulated Author's Rebuttal

3 responses · 0 unresolved

We thank the referee for the careful reading and constructive comments on our manuscript. We address each major comment below, clarifying the arguments in the paper and indicating the revisions we will make to strengthen the exposition and proofs.

read point-by-point responses
  1. Referee: [§3] §3 (height and enumeration): the paper orders curves by the naive height H(E) = max(|c₄|, |Δ|^{1/12}) on minimal Weierstrass models. For the j=0 family the scaling B ↦ u⁶B changes the model while preserving j; the minimality condition at p=3 (v₃(Δ) < 12 after admissible change of variables) restricts the admissible u. It is not shown that the induced distribution on v₃(B) for minimal models still admits the claimed natural density when ordered by H(E); the leading constant in the main asymptotic may therefore be incorrect.

    Authors: We agree that the scaling action and minimality constraint at p=3 must be handled explicitly to confirm the natural density. In the manuscript the curves are enumerated by taking B ∈ ℤ such that y² = x³ + B is already minimal at 3, which restricts v₃(B) to a finite set of residue classes modulo 6. Because the height H(E) is a fixed power of |B|, the count reduces to a one-dimensional lattice-point problem with a periodic local condition at 3. This condition is independent of the global height bound and therefore preserves a positive natural density; the leading constant is simply multiplied by the measure of the admissible classes. We will add a short lemma in §3 that computes this density explicitly and verifies that the leading constant in the main theorems is unaffected. revision: yes

  2. Referee: [Theorem 4.2] Theorem 4.2 (asymptotics for Kodaira symbols at p=3): the claimed proportions (e.g., 1/2 for I₀* and 1/6 for II) are derived from the p-adic valuations of B without an explicit error term or Tauberian argument that accounts for the global height bound. Because the family is thin, the usual sieve or lattice-point counting must be justified; the manuscript provides only a sketch and no verification that the error is o of the main term.

    Authors: The proof of Theorem 4.2 proceeds by partitioning according to v₃(B) = k and applying the local Kodaira–Néron classification for each k; the global count of admissible B with |B| ≪ X is then summed. While the manuscript gives only a sketch, the underlying count is elementary: the number of integers B with |B| ≤ X and v₃(B) = k is (2X)/3^k + O(1), with the error uniform in k up to the height range. Summing the geometric series over k therefore yields an error O(1) for the main term, which is o(X). We will expand the proof of Theorem 4.2 to include this explicit error bound and the short summation argument. revision: partial

  3. Referee: [§5.1] §5.1 (conditional statistics): when the torsion subgroup is fixed, the same height ordering is used, yet the torsion condition imposes congruence conditions on B (or A) that may correlate with v₃(B) (or v₂(A)). The paper does not prove that these correlations are negligible in the asymptotic; a separate major-term calculation or explicit example for the 3-torsion case would be needed.

    Authors: When the torsion subgroup is fixed by a congruence condition modulo m with 3 ∤ m, the Chinese Remainder Theorem makes the condition on B mod m independent of v₃(B), so the joint density factors and the Kodaira proportions remain unchanged. For the 3-torsion case the condition is modulo 9 and does interact with v₃(B). We will add a short explicit computation in §5.1 that enumerates the admissible residue classes modulo 9, recomputes the weighted densities for each Kodaira symbol, and shows that the leading constants are modified by a computable factor (which we record). This supplies the requested major-term calculation and the concrete 3-torsion example. revision: yes

Circularity Check

0 steps flagged

No circularity: standard height counts yield independent asymptotics

full rationale

The paper counts elliptic curves with fixed j-invariants by naive height and derives Kodaira-Néron type asymptotics at the forced bad prime. No quoted step reduces a prediction to a fitted parameter, self-citation chain, or definitional tautology. The derivation uses the explicit Weierstrass models (y²=x³+B for j=0; y²=x³+Ax for j=1728) and standard minimality conditions; these are external to the target proportions and do not collapse by construction. The central claims remain falsifiable against independent enumeration data.

Axiom & Free-Parameter Ledger

0 free parameters · 2 axioms · 0 invented entities

Abstract-only review limits detail; paper appears to rest on domain assumptions about elliptic curve reduction types for fixed j-invariants.

axioms (2)
  • domain assumption Elliptic curves with j-invariant 0 have bad reduction at p=3
    Stated in abstract as background.
  • domain assumption Elliptic curves with j-invariant 1728 have bad reduction at p=2
    Stated in abstract.

pith-pipeline@v0.9.0 · 5409 in / 1111 out tokens · 34008 ms · 2026-05-15T02:45:30.838978+00:00 · methodology

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Lean theorems connected to this paper

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

  • Cost.FunctionalEquation washburn_uniqueness_aczel unclear
    ?
    unclear

    Relation between the paper passage and the cited Recognition theorem.

    We count elliptic curves with j-invariant 0 and 1728 by height and determine asymptotics for the various Kodaira-Néron types at 3 and 2, respectively... N_0_T(X) = c_T · (62/63)√27 ζ(6) X^{1/2} + O(X^{1/4})

  • Foundation.DimensionForcing alexander_duality_circle_linking unclear
    ?
    unclear

    Relation between the paper passage and the cited Recognition theorem.

    E is minimal iff A is fourth-power-free... typ_2(E) given in Tables 6-9

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.

Reference graph

Works this paper leans on

10 extracted references · 10 canonical work pages

  1. [1]

    Barrios, M

    A. Barrios, M. Roy. Local data of rational elliptic curves with nontrivial torsion, Pacific J. Math.318no. 1, 1-42 (2022)

  2. [2]

    Barrios, M

    A. Barrios, M. Roy. Representations attached to elliptic curves with a non-trivial odd torsion point, Bull. Lond. Math. Soc.54no. 5, 1846-1861 (2022)

  3. [3]

    R. Brown. The natural density of some sets of square-free numbers. Integers 21 (2021), Paper No. A81, 9 pp

  4. [4]

    Chiloyan, A

    G. Chiloyan, A. Lozano-Robledo. A classification of isogeny-torsion graphs ofQ-sogeny classes of elliptic curves. Trans. London Math. Soc.8no. 1, 1–34 (2021)

  5. [5]

    Cullinan, M

    J. Cullinan, M. Kenney, J. Voight. On a probabilistic local-global principle for torsion on elliptic curves. J. Th´ eorie Nombres Bordeaux.34(1) 41-90 (2022)

  6. [6]

    G. Frey. Some remarks concerning points of finite order on elliptic curves over global fields, Ark. Mat.15no. 1, 1–19 (1977)

  7. [7]

    The LMFDB Collaboration, The L-functions and modular forms database, http://www.lmfdb.org, 2025, [Online; accessed 22 November 2025]

  8. [8]

    Pari/GP, version2.17.2, Bordeaux, 2024,http://pari.math.u-bordeaux.fr/

  9. [9]

    Silverman

    J. Silverman. The arithmetic of elliptic curves. Graduate Texts in Mathematics,106. Springer, 2009

  10. [10]

    Silverman

    J. Silverman. Advanced topics in the arithmetic of elliptic curves. Graduate Texts in Mathe- matics,151. Springer, 1999. Department of Mathematics, Bard College, Annandale-On-Hudson, NY 12504, USA Email address:cullinan@bard.edu URL:http://faculty.bard.edu/cullinan/ Department of Mathematics, Bard College, Annandale-On-Hudson, NY 12504, USA Email address:...