{"id":"96e534bf-952f-4449-bb86-18f261d77da2","arxiv_id":"2412.05867","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":5.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":0,"one_line_summary":"For all but finitely many anticyclotomic twists of an equivariant Hecke character over any imaginary quadratic field, the central vanishing order of the L-function is 0 or 1 exactly according to the root number.","lead":"This paper extends a 1984 theorem of Rohrlich: for anticyclotomic twists of Hecke characters over imaginary quadratic fields, the L-function almost always vanishes at the central point to the order forced by its sign. The extension drops Rohrlich's class number 1 restriction, so the result now covers all imaginary quadratic fields, a useful step for Birch-Swinnerton-Dyer and Iwasawa-type questions.","discovery_kind":"extension","skeptic_critique":{"model":"deepseek-v4-flash","headline":"For W(χ)=-1 the proof needs L'(1,χ)=0 to imply L'(1,χ^σ)=0 for every K-automorphism σ; this is attributed to Gross-Zagier [3], but [3] is an elliptic-curve statement and the characters here correspond to higher-dimensional GL2-type abelian varieties when h_K>1.","rationale":"I read the paper in good faith; it is a serious generalization of Rohrlich, and the high-level architecture is coherent. The main theorem is conditional on two external inputs: Shimura algebraicity for the v=0 case and a Gross-Zagier-type formula for the v=1 case. The first is a known theorem and applies. The second is the weakest point because the cited source does not match the level of generality. The paper's own introduction acknowledges that the abelian varieties are of GL2-type and generally not elliptic curves when h_K>1, so [3] cannot be the whole story. This is a gap in verification, not a claim that is mathematically false; the missing ingredient (YZZ) is likely available. I therefore do not change the reader's conditional verdict. I also note the internal issues (Proposition 5 modulus typo; Lemma 2 valuation steps), but they are less load-bearing because they are local and repairable. No ad hominem: the concern is about the argument, not the author.","tokens_in":15583,"tokens_out":25386,"duration_ms":282015,"concrete_test":"Take K=Q(√-5) (class number 2), a split rational prime p∈P, and a character χ∈X with conductor p^m and W(χ)=-1. For the newform f_χ attached to χ, check the hypotheses of Yuan-Zhang-Zhang [18] needed to conclude L'(1,χ)=0 ⇒ the relevant Heegner point is torsion: existence of a quaternion algebra B and a K-rational CM point of conductor f(χ) on the corresponding Shimura curve. If these hypotheses fail for this χ or for its conjugates, the implication used at (2.19) is not established and the v=1 half of Theorem 1 lacks proof.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The averaging argument in §2 is sound only if, for every χ∈X and every K-automorphism σ of C, a zero at s=1 is preserved under conjugation: L(1,χ)=0 ⇒ L(1,χ^σ)=0, and W(χ)=-1 with L'(1,χ)=0 ⇒ L'(1,χ^σ)=0 (equations (2.13)-(2.20)). The first is Shimura algebraicity and is used correctly. The second is the load-bearing point. The paper cites [3], the Gross-Zagier CRAS announcement, but [3] is concerned with elliptic curves over Q. For h_K>1, the Hecke character χ is attached not to an elliptic curve but to an abelian variety of GL2-type over Q of dimension h_K, as the Introduction itself says: the result is a 'generalization from elliptic curves to this special kind of abelian varieties'. The derivative-preservation then needs the Shimura-curve Gross-Zagier theorem in the form of Yuan-Zhang-Zhang [18], including a quaternionic setup and a Heegner-type condition on each conductor f(χ). The paper neither states this theorem nor verifies its hypotheses for every χ in the family; it also does not justify that the zero of the height in the formula is Galois-invariant. If the implication fails for some conjugate, L'(1,χ)_av need not vanish when a single conjugate has a zero, and the contrapositive proof of Theorem 1 for W(χ)=-1 collapses. A secondary but related rigor issue is Proposition 5's statement, where the congruence is written modulo Z_p instead of modulo qZ_p, and Lemma 2's steps (3.11)-(3.16) require unstated p-adic valuation arguments; these are local and repairable, unlike the missing Gross-Zagier hypothesis.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper proves that, for all but finitely many Hecke characters χ in a prescribed anticyclotomic family over an arbitrary imaginary quadratic field K, the central vanishing order of L(s,χ) is determined by the root number: ord_{s=1} L(s,χ)=0 when W(χ)=1 and 1 when W(χ)=-1. The family consists of twists φρ where φ has infinite type (1,0) and equivariant under complex conjugation, and ρ is a finite-order anticyclotomic character unramified outside a fixed finite set P. The proof follows Rohrlich's strategy: express L^{(v)}(1,χ) as a sum of principal and non-principal terms, average over K-automorphisms, show the non-principal average is negligible using counting estimates, and invoke p-adic Roth-type theorems. The paper also proves the asymptotics L(1,χ)_av → 2L(1,κ) for W(χ)=1 and L'(1,χ)_av ∼ 2L(1,κ) log(Af) for W(χ)=-1.","tokens_in":15903,"tokens_out":10070,"duration_ms":95475,"significance":"If correct, the theorem is a substantial generalization of Rohrlich's class-number-one result to arbitrary imaginary quadratic fields, replacing elliptic curves with abelian varieties of GL2-type having CM by K. The analytic core—the Abel-theorem asymptotics, the split into principal and non-principal terms, and the exponent bookkeeping in Propositions 1–2—is sound and carefully executed. The novel reduction via h-th roots in Lemma 2 and the passage to Ridout's theorem are interesting technical contributions. The manuscript is not accompanied by machine-checked proofs, but the argument is written in a checkable style and the main external inputs are clearly identified.","major_comments":[{"comment":"The contrapositive step of the averaging argument requires that for every K-automorphism σ of C, if W(χ)=-1 and L'(1,χ)=0, then L'(1,χ^σ)=0. This is load-bearing: without it, a zero of L'(1,χ) need not force the average L'(1,χ)_av to vanish, and the proof of Theorem 1 for W(χ)=-1 collapses. The manuscript attributes this implication to Gross–Zagier [3], but [3] is an elliptic-curve statement. In the present setting, when h_K>1, the Hecke character χ is attached to an abelian variety of GL2-type of dimension h_K, as the Introduction itself states. What is needed is the Shimura-curve Gross–Zagier formula in the form of Yuan–Zhang–Zhang [18], together with a verification of its Heegner-type hypothesis for every χ in the family. The paper neither states this theorem nor checks its hypotheses, and it does not explain why the relevant height vanishing is Galois-invariant. This gap must be repaired before the central claim can be accepted.","section":"Introduction; §2, (2.19)–(2.20)"},{"comment":"The proof of Lemma 2 is too compressed at the point where the condition on the order of ε_p(w^h w_0) is converted into the congruence w ≡ η_p ω_p x_p mod qO⊗Z_p. Specifically, the existence of m=jp^μ satisfying (3.11) from the p^μ-annihilation assumption, the m-th root extraction in (3.12)–(3.13), and the step from z_1=z_2^h to (3.17)–(3.18) all require p-adic valuation estimates that are not stated. Since Lemma 2 is the bridge to the counting propositions and is used for both Propositions 1 and 2, the exponent bookkeeping and the choice of μ should be written out in full.","section":"§3, Lemma 2, (3.11)–(3.16)"}],"minor_comments":[{"comment":"In the statement of Proposition 5, condition (1) says 'uβ_p − vα_p ≡ 0 mod Z_p', but from the proof and from the earlier definition of M(q,t) the congruence should be modulo qZ_p. In addition, line (4.9) writes max(|u|,|v|) ≤ q^{-1}; this should be ≤ q^c, since the set M(q,q^c) is being counted.","section":"§4, Proposition 5"},{"comment":"In the v=1 case, the bound for the third sum is written with e^{Na/Af}; from (2.38) it should be e^{-Na/Af}.","section":"§2, (2.43)"},{"comment":"The set N(χ,t) is defined twice: first with conditions (1)–(3) and then again with an additional condition (4) fixing an ideal class. To avoid confusion, the final definition should be stated once and used consistently from that point on.","section":"§3, definitions of N(χ,t)"},{"comment":"There are several typographical errors and OCR-style artifacts (e.g., 'satifies', 'infinte', missing overlines in condition (2) of N_ω(q,t)); the manuscript would benefit from a careful proofreading pass.","section":"Throughout"}],"recommendation":"major_revision","confidential_remarks":null},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Here is my read. The paper does what it says: it removes the class number 1 assumption from Rohrlich's 1984 anticyclotomic vanishing theorem, and the technical apparatus for doing so is real. The main theorem (vanishing order 0 or 1 according to root number, outside a finite exceptional set) is new for h>1. The h-th power trick — encoding each non-principal ideal a=wa0 through a^h = w^h w0 O, so that χ(a) becomes a root of unity times a fixed h-th root — is a genuine novelty, and the reduction to a counting problem for pairs of integers satisfying p-adic congruences is clean. The Section 2 asymptotics (Abel summation, the v=1 estimate (2.40)-(2.43)) check out. The paper is honest about what it borrows from Rohrlich [9].\n\nThe soft spots. The load-bearing input for the W(χ)=-1 case is the implication L'(1,χ)=0 ⇒ L'(1,χ^σ)=0 for every K-automorphism σ. The paper cites the Gross-Zagier CRAS note [3], but that note is about elliptic curves. For h>1 the Hecke character is attached to a GL2-type abelian variety of dimension h, and the implication needs the Shimura-curve Gross-Zagier formula (Yuan-Zhang-Zhang [18]) plus a Heegner-type condition on each conductor f(χ). The paper neither states the theorem nor verifies the hypotheses. This is not a cosmetic gap: if the implication fails for some conjugate, then a zero of one conjugate does not force the average to vanish, and the contrapositive proof collapses. I think the implication is true in this setting, so the gap is likely repairable, but it must be written out.\n\nTwo smaller issues: Lemma 2's steps (3.11)-(3.16) are very compressed; the p-adic valuation arguments are plausible but unstated. And Section 4 has local typos: Proposition 5 says mod Z_p where it should be mod qZ_p, and (4.9) writes q^{-1} where the argument requires q^c. Neither affects the method.\n\nVerdict: this deserves a serious referee, not a desk reject. The main theorem is the expected generalization, the analytic spine is sound, and the new counting lemma is the right kind of technical content. But the current text does not fully prove the W(χ)=-1 case for h>1 because of the missing Gross-Zagier hypothesis. I would send it out with an explicit request to check that input.","headline":"A real and expected generalization of Rohrlich, with a solid analytic core but a missing Gross-Zagier check for the derivative case when h>1.","tokens_in":16525,"tokens_out":9732,"would_cite":false,"duration_ms":93660,"reading_group":"maybe","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":["11R42","11G40","11M41","11J68"],"pacs":[],"model":"deepseek-v4-flash","headline":"For almost all anticyclotomic twists, the central zero order of the Hecke L-function is exactly the sign-allowed minimum.","keywords":["Hecke L-functions","anticyclotomic extensions","central vanishing order","root numbers","imaginary quadratic fields","Galois conjugation","complex multiplication","p-adic Diophantine approximation"],"falsifier":"Find infinitely many characters χn in X with conductors tending to infinity such that W(χn)=1 but ord_{s=1}L(s,χn)>0, or W(χn)=-1 but ord_{s=1}L(s,χn)>1; the theorem predicts both fail eventually, so the first such sequence would refute it.","tokens_in":15194,"feed_emoji":"🔢","tokens_out":13486,"duration_ms":115041,"temperature":0.7,"pith_summary":"Let $K$ be an imaginary quadratic field and $\\varphi$ a Hecke character of infinite type $(1,0)$ that is equivariant under complex conjugation. The paper studies the anticyclotomic twists of $\\varphi$ whose ramification lies in a fixed finite set of primes, an infinite family whose Galois action is inverted by complex conjugation. It proves that for all but finitely many twists $\\chi$, the Hecke $L$-function $L(s,\\chi)$ vanishes at the central point $s=1$ to order $0$ when the root number $W(\\chi)=1$, and to order $1$ when $W(\\chi)=-1$; here the root number is the sign in the functional equation, so this is exactly the minimal order that parity allows. The result generalizes an earlier theorem for elliptic curves with complex multiplication to the broader class of CM abelian varieties over the rationals, and works for every imaginary quadratic field, not just those with class number one.","feed_headline":"Almost all anticyclotomic twists: L-zero order equals sign","feed_subtitle":"Extends the elliptic-curve theorem to all imaginary quadratic fields, with only finitely many exceptions.","key_machinery":"The argument rests on an averaging identity over the Galois conjugates of $\\chi$. With $v\\in\\{0,1\\}$ fixed by $W(\\chi)=(-1)^v$, one studies $L^{(v)}(1,\\chi)_{\\mathrm{av}}=[K(\\chi):K]^{-1}\\sum_\\sigma L^{(v)}(1,\\chi^\\sigma)$; a zero of one conjugate forces all conjugates to share the zero, so vanishing of the average is necessary for an individual zero. The paper then shows the average is dominated by an explicit principal term $2\\sum_n \\kappa(n)n^{-1}I_v(n^2/A f)$, whose limit is $2L(1,\\kappa)\\neq 0$ in the even case and whose leading growth is $2L(1,\\kappa)\\log(A f)$ in the odd case, with the remaining non-principal contribution sent to zero by a counting argument. The counting is successively reduced to counting pairs of rational integers $(u,v)$ lying in prescribed residue classes modulo $q$, and a p-adic Diophantine approximation bound gives the two key estimates: no such pairs exist up to $q^c$ for $c<1/2$, and fewer than $q^s$, $s<1/2$, exist up to $q^d$ for some $d>1/2$.","core_discovery":"The central claim is that the root number determines the central vanishing order throughout the anticyclotomic tower, up to a finite set of characters. The paper proves that the average of $L^{(v)}(1,\\chi)$ over Galois conjugates tends to $2L(1,\\kappa)$ in the even case and grows like $2L(1,\\kappa)\\log(A f(\\chi))$ in the odd case, where $\\kappa$ is the quadratic Dirichlet character attached to $K$ and $A$ is a constant depending only on $K$. Since a zero of any conjugate forces the average to vanish, the nonvanishing of the average for sufficiently large conductors forces the individual $L$-functions to have the minimal zero order allowed by parity.","pith_inferences":["The finite exceptional set is non-effective because it inherits ineffectivity from the p-adic Diophantine approximation input; an effective version of that bound would turn the theorem into a finite verification.","The same averaging-and-counting strategy may apply to cyclotomic towers or to CM fields of higher degree, provided the two conjugation-inheritance properties and the counting reduction survive.","A direct numerical probe of the first few thousand twists of a fixed base field could test the predicted sign-versus-zero pattern and help locate the exceptional set in practice."],"forward_implications":["For every imaginary quadratic field, the root number determines the central vanishing order for all anticyclotomic twists outside a finite set.","The averaged value $L(1,\\chi)_{\\mathrm{av}}$ tends to $2L(1,\\kappa)\\neq 0$ in the even case, so the family average itself is nonzero.","In the odd case $L'(1,\\chi)_{\\mathrm{av}}$ grows like $2L(1,\\kappa)\\log(A f)$, so simple zeros are generic.","The earlier class-number-one restriction is removed, so the result covers all imaginary quadratic fields."],"supporting_citations":[{"why":"the earlier result for elliptic curves with complex multiplication whose proof pattern the paper follows","marker":"[9]"},{"why":"the algebraicity of special values used to transfer a zero from L(1,χ) to every conjugate L(1,χσ)","marker":"[15]"},{"why":"the companion period result that completes the same transfer for critical special values","marker":"[16]"},{"why":"the derivative special-value formula invoked to transfer L'(1,χ)=0 when W(χ)=-1","marker":"[3]"},{"why":"the full derivative formula for arbitrary conductors that the derivative transfer requires","marker":"[18]"},{"why":"the p-adic Diophantine approximation bound that makes the counting estimates work","marker":"[8]"},{"why":"the explicit root number formula used to prove W(χσ)=W(χ)","marker":"[7]"}],"fun_headline_variants":["Root number determines L-zero order for almost all anticyclotomic twists","Zero order from root number: almost all anticyclotomic twists","Root number fixes L-zero order for almost all anticyclotomic twists","Almost all anticyclotomic twists: minimal zero order from parity","Parity dictates L-zero order for all but finitely many twists"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The proof depends on the implication that a zero of L(1,χ), or of L'(1,χ) when W(χ)=-1, forces the same zero for every Galois conjugate χσ; in the derivative case this uses a special-value formula for CM points that the paper assumes applies to every conductor in the family.","fun_headline_variants_meta":{"raw":{"variants":["Root number determines L-zero order for almost all anticyclotomic twists","Zero order from root number: almost all anticyclotomic twists","Root number fixes L-zero order for almost all anticyclotomic twists","Almost all anticyclotomic twists: minimal zero order from parity","Parity dictates L-zero order for all but finitely many twists"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000702,"raw_usage":{"total_tokens":3106,"prompt_tokens":822,"completion_tokens":2284,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":438,"completion_tokens_details":{"reasoning_tokens":2187}},"tokens_in":438,"tokens_out":2284,"duration_ms":14360,"temperature":1.0,"reasoning_tokens":2187,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-11T20:20:50.197374+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Find infinitely many characters χn in X with conductors tending to infinity such that W(χn)=1 but ord_{s=1}L(s,χn)>0, or W(χn)=-1 but ord_{s=1}L(s,χn)>1; the theorem predicts both fail eventually, so the first such sequence would refute it.","supporting_citations":[{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"the earlier result for elliptic curves with complex multiplication whose proof pattern the paper follows"},{"cited_title":"Shimura, The special values of the zeta functions associated with cus p forms , Comm","cited_arxiv_id":null,"evidence_quote":"the algebraicity of special values used to transfer a zero from L(1,χ) to every conjugate L(1,χσ)"},{"cited_title":"Shimura, On the periods of modular forms , Math","cited_arxiv_id":null,"evidence_quote":"the companion period result that completes the same transfer for critical special values"},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"the derivative special-value formula invoked to transfer L'(1,χ)=0 when W(χ)=-1"},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"the full derivative formula for arbitrary conductors that the derivative transfer requires"},{"cited_title":"Ridout, The p-adic generalization of the Thue-Siegel-Roth theorem , Mathematika, 5(1958), 40-48","cited_arxiv_id":null,"evidence_quote":"the p-adic Diophantine approximation bound that makes the counting estimates work"}],"review_version":1}