REVIEW 3 major objections 5 minor 1 cited by
Hilbert Eisenstein series as Doi-Naganuma lift
T0 review · 3 major / 5 minor · reviewed 2026-08-07 · deepseek-v4-flash
Pith's one-line read For real quadratic fields with Galois CM, every incoherent Hilbert Eisenstein series is a Doi–Naganuma lift of a rational Eisenstein series.
desk verdict Substantial, likely correct, and fills a real gap; the only serious worry is the compressed local surjectivity proof in Prop 4.10. read the letter →
The pith
A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.
The reading
What carries the argument
The machinery is the Doi–Naganuma $\theta$ integral $I(h,\Xi,\varphi,s)=\int_{G(\mathbb{Q})\backslash G(\mathbb{A})} E(g,\Xi,s)\,\theta_V(g,h,\varphi)\,dg$, which unfolds to a sum over $B(F)\backslash G(F)$ of local sections $F(h,s;\Xi,\varphi)$ as in (2.32). The global matching theorem is assembled from a standard archimedean identity, the spherical matching at unramified primes, and a surjectivity statement (Proposition 4.10) for the map $(\Xi,\varphi)\mapsto F(\cdot,0;\Xi,\varphi)$ from $I(0,\chi_{2,p})\times S(V(\mathbb{Q}_p))$ into $I(0,\chi_p)$. Surjectivity is detected through Whittaker coefficients: a suitable coefficient is shown to be nonzero for every class $\alpha\in F_p^\times/\mathrm{Nm}(E_p^\times)$, and a lemma on simple modules then forces the image to be the whole space. The explicit examples use new fundamental invariant vectors $u_K$ in the finite Weil representation, obtained as the one-dimensional $\chi$-isotypic part of the group ring.
What would settle it
At a ramified prime $p$ (especially $p=2$), compute the explicit Whittaker coefficient $W_m(F_p(\cdot,0;\Xi_p,\varphi))$ for every class $m\in\mathcal{O}_p$ required by Proposition 4.10; a single zero where the proof needs a nonzero value would break the local surjectivity and with it Theorems 1.1 and 1.6.
Extended reading notes
Core claim
The central claim is Theorem 1.1: for any incoherent Siegel section $\Phi \in I(s,\chi)$ attached to a CM extension $E/F$ with $E/\mathbb{Q}$ Galois, there are a standard rational Eisenstein section $\Xi \in I(s,\chi_2)$ and a Schwartz function $\varphi\in S(V(\mathbb{A}))$ such that $I(h,\Xi,\varphi,s)=C(s)E(h,\Phi,s)+O(s^n)$ for a real-analytic $C(s)$ and any $n\ge 1$. In the simplest level-one case the identity is exact, with $C(s)$ supplied by the completed $L$-function $\Lambda(s+1,\chi_1)$. The paper proves this by showing the local matching map is surjective at every finite prime, and it records explicit matching data in terms of new invariant vectors in the Weil representation. It then substitutes the $\theta$-integral expression into the CM-value formula for higher Green functions to compute the generalized Rankin–Selberg $L$-function and to derive the non-unit result for Borcherds products.
Load-bearing premise
The proof stands on the local assertion that for every ramified prime the map $(\Xi,\varphi)\mapsto F(\cdot,0;\Xi,\varphi)$ is surjective onto $I(0,\chi_p)$, which hinges on a nonvanishing calculation of Whittaker coefficients whose $p=2$ cases are deferred.
Editorial extensions
If this is right
- For real quadratic fields with quartic Galois CM, every incoherent Hilbert Eisenstein series is, up to $O(s^n)$, a Doi–Naganuma lift, so its diagonal restriction inherits a spectral expansion from a rational Eisenstein series.
- The generalized Rankin–Selberg $L$-function $L(s,G;D_1,D_2)$ studied in [11] is computed explicitly in terms of $L(G,s+k)$, partial zeta values, and the Fourier coefficients $c_{\tilde G}(|D_1|)c_{\tilde G}(|D_2|)$ of the Shintani lift; its vanishing is controlled by the Fricke eigenvalue $\epsilon(G)$.
- For odd square-free $N$, the value $\Psi(\tau_1,\tau_2)$ of a Borcherds product on $X_0(N)^2$ with effective divisor is not an integral unit when $\max(|D_1|,|D_2|)$ is sufficiently large, extending the level-one singular-moduli result.
- The new invariant vectors $u_K$ give explicit Schwartz functions at ramified primes, making the matching identity exact in the level-one case and computable in genus-zero examples such as differences of hauptmoduls.
Reading between the lines
- A natural next step is to complete the deferred $p=2$ cases of Lemma 5.1; doing so should remove the odd-square-free technical condition on $N$ in Theorem 1.6, as the paper itself anticipates.
- The one-dimensionality of the $\chi$-isotypic part of the finite Weil representation suggests that the same invariant-vector construction works for other ramified quadratic extensions and finite quadratic modules, yielding explicit local matching data beyond the quartic-Galois setting.
- The explicit formula for $L(s,G;D_1,D_2)$ can be tested numerically for small $D_1,D_2$ and $N$ by computing both sides from Fourier expansions, which would also probe the conjectural arithmetic intersection interpretation attached to this $L$-function in [11].
Editorial analysis
A structured set of objections, weighed in public.
Referee Report
Summary. The paper proves that incoherent Hilbert Eisenstein series for a real quadratic field F, associated with a CM extension E/F with E/Q Galois, can be expressed as the Doi-Naganuma theta lift of an incoherent Eisenstein series over Q (Theorem 1.1). The proof proceeds by unfolding the theta integral to a local matching problem and solving it locally at archimedean, unramified, and ramified places, with the ramified case reduced to a local surjectivity statement (Proposition 4.10). An explicit version is given for biquadratic fields (Theorem 1.2), yielding a formula for a generalized Rankin-Selberg L-function (Theorems 1.4 and 6.4). This is applied to prove that Borcherds products on X_0(N)^2 with effective divisors are not integral units at Heegner points of sufficiently large discriminants (Theorem 1.6), generalizing earlier work of the first author. The paper also contains a new construction of fundamental invariant vectors in the Weil representation of finite quadratic modules (Section 5.1).
Significance. If correct, the main theorem is a significant structural result: it realizes level-one incoherent Hilbert Eisenstein series as Doi-Naganuma lifts, filling a gap for small and non-parallel weight. The explicit local matching and the resulting Rankin-Selberg formula are valuable and give a concrete handle on L-functions that appeared in the work of Bruinier-Kudla-Yang. The non-integral-unit application generalizes a known result for singular moduli to higher level and is a natural target of the method. The paper is largely explicit and builds on standard machinery; the main risk is the compressed local nonvanishing computation and a deferred p=2 case, as detailed below.
major comments (3)
- [4.3 (Proposition 4.10)] The nonvanishing computation is load-bearing for Theorem 4.1 and hence Theorem 1.1, but it is not fully proved. The invariance of φ under J_p∩N_p^- is asserted after a one-sentence Fourier-transform computation, and the normalization of the additive measure and of the volume of J_p∩N_p^- in the definition of C_1(φ) is not stated. In particular, the displayed equality C_1(φ)=φ in the proof should be C_1(φ)=vol(J_p∩N_p^-)φ up to a Weil-index scalar; this does not affect the nonvanishing conclusion, but it must be written correctly. Please expand the computation, including the split case and the action of the conjugate subgroup, so that the surjectivity claim is checkable by the reader.
- [5.1 (Lemma 5.1) and 6.2] Lemma 5.1 defers the p=2 case with 'the argument can be suitably adapted' and footnote 3 'Alternatively, one can check all cases by hand.' Theorem 6.4 is stated for D_1,D_2 not both even, and the construction of φ_{D_1,N_1,N_2} in (6.2) invokes u_{Q_p(√D_1)} for p|D_1; when D_1 is even, this requires exactly the p=2 case of Lemma 5.1. Since Theorem 1.6 relies on Theorem 6.4, the p=2 gap is load-bearing for the full statement of Theorem 1.6. Either prove the p=2 case, or restrict Theorem 6.4 to odd D_1 and explain, using the symmetry of L(s,G;D_1,D_2), why Theorem 1.6 follows for all allowed D_1,D_2.
- [4.4 (Proof of Theorem 4.1)] The proof says 'we denote the set of the other places by S, which is non-empty.' For a Siegel section that is spherical at every finite place, S is empty and the argument does not cover arbitrary C(s). The statement of Theorem 4.1 does not restrict to incoherent sections. Please either restrict the statement to the incoherent setting, where the finite ramified places are naturally present, or modify the proof so that when S is empty one applies Proposition 4.10 at an arbitrary finite prime.
minor comments (5)
- [4.3] In Proposition 4.10, the phrase 'When F_p=Q_p^2, let them be Z_p^2' for O_p and d_p^{-1} is confusing; the split case should be written explicitly in terms of the product structure F_p=Q_p×Q_p.
- [4.3] The measure on J_p∩N_p^- in (4.7) is not normalized; since C_1(φ) is defined without a normalization factor, the equality C_1(φ)=φ in the proof of Proposition 4.10 is dimensionally incorrect and should be corrected.
- [References] Reference [14] and [15] are the same Duke article and should be merged.
- [6.3 (Proof of Theorem 1.6)] The estimate |L'(0,ξ(f);D_1,D_2)| ≪_N |D_1D_2|^{-δ} is asserted from [15]; please state the precise Fourier-coefficient bound used and the resulting δ, since the standard convexity bound would not suffice.
- [6.2 (Lemma 6.3)] In (6.17), the notation Im(R_{H_0}) should be defined; presumably it denotes the image of the raising operator on functions on H_0(R).
Circularity Check
No circular derivation: the local-to-global matching proof is self-contained; self-citations are antecedents, and the deferred p=2 computation is a completeness gap, not circularity.
full rationale
The claimed derivation chain is a local-to-global matching theorem. Theorem 1.1 is obtained by unfolding the theta integral (Lemma 2.4) and proving the local matching statement Theorem 4.1. No parameter is fitted to the quantity being 'predicted': the constants C(s), Gamma_R, and L(s, chi_1) are canonical, and the matching data (Xi, phi) are constructed for every Siegel section Phi rather than calibrated to a subset of its coefficients. The central surjectivity Proposition 4.10 is proved from the Kudla-Rallis decomposition of I(0, chi_p) (external reference [31]) plus an explicit Whittaker-coefficient computation; the nonvanishing is not imported from a self-citation. Self-citations [33] and [34] are antecedents being generalized and enter as technique, not as load-bearing premises; [9] supplies a theta-lift identification and Fock-model notation but does not posit Theorem 1.1. The review-rule flags in the manuscript are completeness gaps rather than circular steps: Proposition 4.10's ramified-prime Whittaker computation is compressed, and Lemma 5.1 explicitly defers p=2 ('the argument can be suitably adapted for p=2'); if those fail, the derived theorems would not follow, but that is a soundness risk, not a reduction of the conclusion to the hypotheses. I therefore find no significant circularity.
Assumptions & free parameters
assumptions (5)
- standard math Kudla-Rallis decomposition of I(0,chi_p) into irreducible summands R(W_alpha) [31]
- standard math Siegel-Weil embedding and local theta correspondence for quadratic chi2
- domain assumption CM-value formula for higher Green functions at big CM cycles [10, Theorem 5.10]
- domain assumption Equidistribution of CM points on X0(N)^2 (Duke [14]) and Brauer-Siegel lower bound for class numbers
- domain assumption Subconvexity bounds for GL1-twists of GL2 forms and non-trivial bounds for half-integral weight Fourier coefficients
Cite this review
Pith. "Pith review of Hilbert Eisenstein series as Doi-Naganuma lift." pith.science (2026). https://pith.science/paper/EL7CZTUK
@misc{pith2026250601688,
author = {Pith},
title = {Pith review of: Hilbert Eisenstein series as Doi-Naganuma lift},
year = {2026},
howpublished = {\url{https://pith.science/paper/EL7CZTUK}},
note = {Machine review of arXiv:2506.01688}
}
abstract
In this paper, we show that incoherent Hilbert Eisenstein series for a real quadratic fields can be expressed as the Doi-Naganums lift of an incoherent Eisenstein series over $\mathbb{Q}$. As an application, we show when $N$ is odd and square-free, the values at Heegner points of Borcherds product on $X_0(N)^2$ with effective divisors are not integral units when the discriminants are sufficiently large. This generalizes a result of the first author to higher levels. In the process, we explicitly describe the Rankin-Selberg type L-function that appeared in the work of Bruinier-Kudla-Yang when the quadratic space has signature (2, 2), and give a new construction of fundamental invariant vectors appearing in Weil representations of finite quadratic modules.
Forward citations
Cited by 1 Pith paper
-
Twisted Siegel-Weil formulas for $\mathrm{GL}_2$ over non-Galois quartic CM fields
Twisted theta integrals over non-Galois quartic CM fields equal Doi–Naganuma lifts of Hecke's integral, making twisted CM values of Borcherds forms algebraic multiples of logarithms of units.
Reference graph
Works this paper leans on
-
[1]
Claudia Alfes-Neumann and Markus Schwagenscheidt,Shintani theta lifts of harmonic Maass forms, Trans. Amer. Math. Soc.374(2021), no. 4, 2297–2339. MR 4223017 30, 32
work page 2021
-
[2]
Fabrizio Andreatta, Eyal Z. Goren, Benjamin Howard, and Keerthi Madapusi Pera,Faltings heights of abelian varieties with complex multiplication, Ann. of Math. (2)187(2018), no. 2, 391–531. MR 3744856 3
work page 2018
-
[3]
Number Theory250(2023), 155–182
Patrick Bieker,Invariants for the Weil representation and modular units for orthogonal groups of signature(2,2), J. Number Theory250(2023), 155–182. MR 4584422 5
work page 2023
-
[4]
Yuri Bilu, Philipp Habegger, and Lars K¨ uhne,No singular modulus is a unit, Int. Math. Res. Not. IMRN (2020), no. 24, 10005–10041. MR 4190395 4
work page 2020
-
[5]
Gabriele Bogo and Yingkun Li,Span of restriction of Hilbert theta functions, Pure Appl. Math. Q. 19(2023), no. 1, 61–83. MR 4570157 16 38 Y. LI AND M. ZHANG
work page 2023
-
[6]
Gabriele Bogo, Yingkun Li, and Markus Schwagenscheidt,Laurent expansions of meromorphic mod- ular forms, J. Lond. Math. Soc. (2)110(2024), no. 6, Paper No. e70036, 34. MR 4833342 34
work page 2024
-
[7]
Borcherds,Automorphic forms with singularities on Grassmannians, Invent
Richard E. Borcherds,Automorphic forms with singularities on Grassmannians, Invent. Math.132 (1998), no. 3, 491–562. MR 1625724 (99c:11049) 9, 15
work page 1998
-
[8]
Jan Bruinier and Ken Ono,Heegner divisors,L-functions and harmonic weak Maass forms, Ann. of Math. (2)172(2010), no. 3, 2135–2181. MR 2726107 5
work page 2010
Show all 44 references
-
[9]
Bruinier, Yingkun Li, and Tonghai Yang,Deformations of Theta Integrals and a Conjecture of Gross-Zagier, arxiv:2204.10604, submitted, 2022
Jan H. Bruinier, Yingkun Li, and Tonghai Yang,Deformations of Theta Integrals and a Conjecture of Gross-Zagier, arxiv:2204.10604, submitted, 2022. 5, 14, 15, 17, 33
2022 arXiv
-
[10]
Math.225(2021), no
Jan Hendrik Bruinier, Stephan Ehlen, and Tonghai Yang,CM values of higher automorphic Green functions for orthogonal groups, Invent. Math.225(2021), no. 3, 693–785. MR 4296349 5, 14, 18
2021
-
[11]
Kudla, and Tonghai Yang,Special values of Green functions at big CM points, Int
Jan Hendrik Bruinier, Stephen S. Kudla, and Tonghai Yang,Special values of Green functions at big CM points, Int. Math. Res. Not. IMRN (2012), no. 9, 1917–1967. MR 2920820 1, 3, 10, 29, 37
2012
-
[12]
228, Springer-Verlag, New York, 2005
Fred Diamond and Jerry Shurman,A first course in modular forms, Graduate Texts in Mathematics, vol. 228, Springer-Verlag, New York, 2005. MR 2112196 8
2005
-
[13]
Math.9(1969/70), 1–14
Koji Doi and Hidehisa Naganuma,On the functional equation of certain Dirichlet series, Invent. Math.9(1969/70), 1–14. MR 253990 2
1969
-
[14]
Duke,Hyperbolic distribution problems and half-integral weight Maass forms, Invent
W. Duke,Hyperbolic distribution problems and half-integral weight Maass forms, Invent. Math.92 (1988), no. 1, 73–90. MR 931205 5
1988
-
[15]
Math.92 (1988), no
,Hyperbolic distribution problems and half-integral weight Maass forms, Invent. Math.92 (1988), no. 1, 73–90. MR 931205 37
1988
-
[16]
Stephan Ehlen and Nils-Peter Skoruppa,Computing invariants of the Weil representation, L-functions and automorphic forms, Contrib. Math. Comput. Sci., vol. 10, Springer, Cham, 2017, pp. 81–96. MR 3931449 5
2017
-
[17]
MR 1050763 2
Eberhard Freitag,Hilbert modular forms, Springer-Verlag, Berlin, 1990. MR 1050763 2
1990
-
[18]
Gross, W
B. Gross, W. Kohnen, and D. Zagier,Heegner points and derivatives ofL-series. II, Math. Ann.278 (1987), no. 1-4, 497–562. MR 909238 (89i:11069) 4, 5, 14, 26
1987
-
[19]
Gross and Don B
Benedict H. Gross and Don B. Zagier,On singular moduli, J. Reine Angew. Math.355(1985), 191–
1985
-
[20]
Math.84(1986), no
,Heegner points and derivatives ofL-series, Invent. Math.84(1986), no. 2, 225–320. 2
1986
-
[21]
Number Theory 2(2016), Paper No
Yueke Hu,Cuspidal part of an Eisenstein series restricted to an index 2 subfield, Res. Number Theory 2(2016), Paper No. 33, 61. MR 3575841 3
2016
-
[22]
Math.24(2019), 1–45
Rodney Keaton and Ameya Pitale,Restrictions of Eisenstein series and Rankin-Selberg convolution, Doc. Math.24(2019), 1–45. MR 3935491 3
2019
-
[23]
Reine Angew
Winfried Kohnen,Newforms of half-integral weight, J. Reine Angew. Math.333(1982), 32–72. MR 660784 9, 35
1982
-
[24]
Winfried Kohnen,Fourier coefficients of modular forms of half-integral weight, Mathematische An- nalen271(1985), 237–268. 8, 32
1985
-
[25]
Kristina Kublik,Generalizations of goursat’s theorem for groups, Rose-Hulman Undergraduate Math- ematics Journal11(2010), Issue 1, Articile 6. 21
2010
-
[26]
Kudla,Theta-functions and Hilbert modular forms, Nagoya Math
Stephen S. Kudla,Theta-functions and Hilbert modular forms, Nagoya Math. J.69(1978), 97–106. MR 466025 2
1978
-
[27]
J.86(1997), no
,Algebraic cycles on Shimura varieties of orthogonal type, Duke Math. J.86(1997), no. 1, 39–78. MR 1427845 2
1997
-
[28]
,Central derivatives of Eisenstein series and height pairings, Ann. of Math. (2)146(1997), no. 3, 545–646. MR 1491448 8
1997
-
[29]
,From modular forms to automorphic representations, An introduction to the Langlands pro- gram (Jerusalem, 2001), Birkh¨ auser Boston, Boston, MA, 2003, pp. 133–151. MR 1990378 8
2001
-
[30]
3, 293–349
,Integrals of Borcherds forms, Compositio Math.137(2003), no. 3, 293–349. MR 1988501 9, 30 HILBERT EISENSTEIN SERIES AS DOI-NAGANUMA LIFT 39
2003
-
[31]
Kudla and Stephen Rallis,Ramified degenerate principal series representations forSp(n), Israel J
Stephen S. Kudla and Stephen Rallis,Ramified degenerate principal series representations forSp(n), Israel J. Math.78(1992), no. 2-3, 209–256. MR 1194967 5, 10, 21
1992
-
[32]
Langlands,Base change forGL(2), Annals of Mathematics Studies, vol
Robert P. Langlands,Base change forGL(2), Annals of Mathematics Studies, vol. 96, Princeton University Press, Princeton, N.J.; University of Tokyo Press, Tokyo, 1980. 2
1980
-
[33]
Ann.368(2017), no
Yingkun Li,Restriction of Coherent Hilbert Eisenstein series, Math. Ann.368(2017), no. 1-2, 317–
2017
-
[34]
Math.157(2021), no
,Singular units and isogenies between CM elliptic curves, Compos. Math.157(2021), no. 5, 1022–1035. MR 4251608 1, 4, 5, 14, 17, 29, 37
2021
-
[35]
Math.234(2023), no
,Algebraicity of higher Green functions at a CM point, Invent. Math.234(2023), no. 1, 375–418. MR 4635835 15
2023
-
[36]
Manuel K. H. M¨ uller and Nils R. Scheithauer,The invariants of the weil representation ofSL 2(Z),
-
[37]
Hiroshi Saito,Automorphic forms and algebraic extensions of number fields, Kinokuniya Book Store Co., Ltd., Tokyo, 1975, Department of Mathematics, Kyoto University, Lectures in Mathematics, No
1975
-
[38]
Takuro Shintani,On construction of holomorphic cusp forms of half integral weight, Nagoya Math. J. 58(1975), 83–126. MR 389772 5
1975
-
[39]
,On liftings of holomorphic cusp forms, Automorphic forms, representations andL-functions (Proc. Sympos. Pure Math., Oregon State Univ., Corvallis, Ore., 1977), Part 2, Proc. Sympos. Pure Math., XXXIII, Amer. Math. Soc., Providence, RI, 1979, pp. 97–110. MR 546611 2
1977
-
[40]
Tonghai Yang and Hongbo Yin,Difference of modular functions and their CM value factorization, Trans. Amer. Math. Soc.371(2019), no. 5, 3451–3482. MR 3896118 15, 16, 17
2019
-
[41]
China Math.65(2022), no
Dongxi Ye,Difference of a Hauptmodul forΓ 0(N)and certain Gross-Zagier type CM value formulas, Sci. China Math.65(2022), no. 2, 221–258. MR 4372919 15
2022
-
[42]
Math.30(1975), no
Don Zagier,Modular forms associated to real quadratic fields, Invent. Math.30(1975), no. 1, 1–46. MR 382174 2
1975
-
[43]
Number Theory9 (2023), no
Shaul Zemel,Integral bases and invariant vectors for Weil representations, Res. Number Theory9 (2023), no. 1, Paper No. 5, 27. MR 4517326 5 Max Planck Institute for Mathematics, Vivatsgasse 7, D–53111 Bonn, Germany Email address:yingkun@mpim-bonn.mpg.de F achbereich Mathematik...
2023
-
[220]
MR 772491 (86j:11041) 2, 4
Reviewed August 7, 2026 · model on record in the stance chip above.
Discussion (0). Sign in to comment.