Pith. sign in

REVIEW 1 major objections 4 minor 22 references

Twisted doubling integrals for classical groups

T0 review · 1 major / 4 minor · reviewed 2026-08-14 · deepseek-v4-flash

Pith's one-line read A uniform unfolding argument reduces the twisted doubling integrals for all classical groups, including quaternionic unitary groups, to a single double coset.

desk verdict A serious, well-written systematization of twisted doubling integrals with a real but explicit gap in the quaternionic unitary case: the existence of type (k,n)_D representations is assumed, not proved. read the letter →

arxiv 1908.10298 v2 pith:FWPAP5L7 submitted 2019-08-27 math.NT math.RT

classification math.NTmath.RT MSC 11F7011F5522E5022E55
keywords twisteddoublingintegralsdegenerateWhittakercoefficientsnilpotentorbitsclassicalgroupsquaternionicunitaryEisensteinseriestensorproductL-functionsunfolding
verification ladder T0 review T1 audit T2 compute T3 formal

The pith

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

The reading

This paper rephrases the twisted doubling integrals -- a family of global zeta integrals built to represent tensor-product $L$-functions of classical groups -- in a conceptual form that works uniformly for every classical group in the setup, including quaternionic unitary groups. Its central result is that, for $\operatorname{Re}(s)$ sufficiently large, the full global integral collapses under unfolding to the contribution of a single double coset: the integral over $G^{\lozenge}(F)\setminus(G\times G)(\mathbb{A})$, with the Eisenstein series replaced by its defining section and a degenerate Whittaker integral over $N^\bullet_W\cap P$. Because that residual integral is a $(k,n)_D$-Fourier coefficient of $\theta\cdot\nu^s$, it is Eulerian, so the global zeta integral becomes an explicit product of local integrals. The paper also computes the relevant Fourier coefficient of the Siegel Eisenstein series and proves a local multiplicity-one statement for supercuspidal inputs.

What carries the argument

The central objects are the degenerate Whittaker coefficients attached to flags of totally isotropic subspaces: for a flag $Y$ and a character $\psi_A$ built from the reduced trace, the pair $(N(Y),\psi_A)$ labels a nilpotent orbit, and a representation of type $(k,n)_D$ is one that admits a unique such functional in the orbit $(kn)_D$ and none in higher orbits. That uniqueness produces the stabilizer character $\chi_\theta$ and the factorization of global Fourier coefficients into local ones (Lemmas 2.15 and 2.17). The unfolding is carried by the invariant $\kappa(L)$, which classifies the double cosets that survive the character test; the Eisenstein series is induced from $\theta\cdot\nu^s$ on the parabolic $P(W^{\Delta,k})$ of the doubled group $G^{\square,k}$.

What would settle it

Compute, for a non-Archimedean $F$ and a division quaternion algebra $D$, the dimension of $\operatorname{Hom}_{N(Y)}(\theta,\psi_A)$ for a representation $\theta$ attached to the orbit $(kn)_D$, for example with $k=n=2$. A dimension greater than $1$ would break the factorization in Lemma 2.15 and hence the Euler-product conclusion of Theorem 5.3. A second concrete test is to search for a maximal isotropic subspace $L$ with $\kappa(L)=0$ that is not in the $\iota(G\times G)N^\bullet_W(F)$-orbit of $W^{\Delta,k}$; if one exists, the collapse to a single double coset would fail.

Watch

Extended reading notes

Core claim

The paper establishes an unfolding identity (Theorem 5.3): for an irreducible cuspidal automorphic representation $\pi$ of a classical group $G$, a representation $\theta$ of $GL_{kn,D}$ of type $(k,n)_D$, and a normalized Siegel Eisenstein series $E(\varphi(s))$ on the doubled group $G^{\square,k}$, the twisted doubling integral $Z(\xi_1\boxtimes\xi_2,\varphi(s))$ equals the integral over $G^\lozenge(F)\setminus(G\times G)(\mathbb{A})$ of $\chi_\theta(\nu(g_2))^{-1}\xi_1(g_1)\xi_2(g_2)$ times the degenerate Whittaker integral of $\varphi(s)$ along $N^\bullet_W\cap P$, when $\operatorname{Re}(s)$ is large. The proof eliminates all but one double coset: the degenerate character forces $L\cap Y_{k-1}=\{0\}$; an invariant $\kappa(L)$, defined as the common dimension of $\overline{L}\cap W_{1,+}$ and $\overline{L}\cap W_{1,-}$, classifies the remaining orbits; and any $L$ with $\kappa(L)>0$ produces an inner integral over a nontrivial unipotent subgroup of a cusp form, hence vanishes by cuspidality. Only $L=W^{\Delta,k}$ survives, and the resulting integral is a $(k,n)_D$-Fourier coefficient of $\theta\cdot\nu^s$, decomposing as an Euler product. The same tools give a one-dimensional space of equivariant functionals for the Eisenstein-series Fourier coefficient (Proposition 7.10) and a local multiplicity-one bound for supercuspidal inputs (Proposition 7.11).

Load-bearing premise

The construction needs, at every place, a representation $\theta$ of the general linear group whose Fourier coefficient attached to the orbit $(kn)_D$ is one-dimensional; the paper itself notes that this uniqueness is not automatic when the underlying division algebra is not a field, and the quaternionic extension relies on it.

Editorial extensions

If this is right

  • For every classical group in the setup, the global twisted doubling integral is Eulerian: it equals an explicit product of local integrals indexed by the places of $F$.
  • The construction now covers quaternionic unitary groups, so tensor-product zeta integrals for these groups can be studied by the same unfolding and factorization arguments.
  • The Fourier-coefficient calculation fixes the normalization of intertwining operators needed for the local theory of the twisted doubling integrals.
  • The local multiplicity-one statement gives a route to local $L$-factors and $\varepsilon$-factors for the tensor product, extending the doubling-method program.
  • Specializing $\theta$ to generalized Speh representations makes the global integral represent a tensor-product $L$-function, and isobaric sums of such representations give products of tensor-product $L$-functions.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

  • If the multiplicity-one condition in Definition 2.8 turns out to fail for a division quaternion algebra in some explicit case, the Euler factorization would break, so the actual reach of the quaternionic extension is exactly as wide as that uniqueness holds.
  • The same double-coset collapse is likely to work for covering groups of classical groups, since the elimination steps depend only on orbit geometry and on the $(k,n)_D$ functional.
  • One could test the method with a different inducing representation than $\theta$: any representation with the same degenerate Whittaker uniqueness should yield an Eulerian integral, potentially representing products or quotients of tensor-product $L$-functions.
  • The invariant $\kappa(L)$ may serve as a general tool for computing Fourier coefficients of other Eisenstein series on classical groups, replacing combinatorial double-coset enumeration by a single dimension invariant.
Share X Bluesky LinkedIn Reddit HN

Signed reviews

No signed human review yet.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Referee Report

1 major / 4 minor

Summary. This paper gives a conceptual reformulation of the twisted doubling integrals of Cai–Friedberg–Ginzburg–Kaplan, and claims to extend the construction to quaternionic unitary groups. The author defines a family of degenerate Whittaker coefficients, introduces the notion of representations of type (k,n)_D for GL_{kn,D}, and uses these tools to prove a uniform unfolding identity (Theorem 5.3) expressing the global twisted doubling integral as an integral over a single double coset, together with an Euler product decomposition. The paper also computes a Fourier coefficient of the Siegel Eisenstein series (Theorem 4.1) and states local analogues of the global results (Propositions 7.10 and 7.11) using Bernstein's localization principle.

Significance. Conditionally on the existence of representations of type (k,n)_D, the paper provides a clean and uniform treatment of the unfolding, which is of independent interest for split classical groups. The split-group argument is detailed and follows standard techniques, and the auxiliary results on degenerate Whittaker coefficients are useful. The main weakness is that the claimed extension to quaternionic unitary groups is not established because the existence of type (k,n)_D representations when D is a division quaternion algebra is not proved or precisely referenced; this is a load-bearing input for the main global identity and for the local results.

major comments (1)
  1. [§2.4 (Remarks 2.9 and 2.13)] The claimed extension to quaternionic unitary groups is not established because the manuscript does not prove or directly cite a result ensuring the existence of irreducible automorphic representations of GL_{kn,D}(A) of type (k,n)_D when D is a division quaternion algebra. Remark 2.9 explicitly warns that the implication from the nilpotent orbit condition to multiplicity one is false when D is not a field, so the type (k,n)_D condition is a genuinely restrictive extra assumption. The cited [Gin06, Prop. 5.3], [JL13, Thm. 1], and [CFK18, Thm. 5] are not shown to apply to GL_{kn,D} for quaternionic D; these references appear to concern the field case, and the manuscript does not explain how their results extend to division quaternion algebras. This is load-bearing for Theorem 5.3 and the Euler product factorization in Section 5.2, which rely on Lemmas 2.15–2.17, as well as for the local results Propositions 7.10 and 7.11. The author should supply a proof or a precise reference covering the quaternionic case, or alternatively present the quaternionic extension as conditional on the existence of such θ.
minor comments (4)
  1. [§1 and §2.2.2] The abstract and introduction state that the construction extends to "all classical groups," but the formal setup assumes W admits a complete polarization (Section 2.2.2). This should be qualified to avoid overclaiming the scope.
  2. [Lemma 2.15] The factorization λ(φ) = ∏_v λ_v(φ_v) is asserted with the sentence "From Definition 2.8, one can show that"; since this factorization is a key step for the Eulerian property, a brief justification would improve readability.
  3. [Remark 2.12] The statement "seems to be redundant (but we cannot find a reference for this)" is informal for a published paper; either provide a proof of redundancy or phrase the remark differently.
  4. [§2.1] The sentence "we fix a nontrivial additive character character ψ_F of F" contains a duplicated word.

Circularity Check

1 steps flagged · score 4.0 of 10

The unfolding argument itself is independent, but the advertised quaternionic unitary extension rests on a load-bearing self-cited existence theorem for type-(k,n)_D representations.

  1. self citation load bearing [Remark 2.13, with Remark 2.9; input to Theorem 5.3 and Section 5.2]
    "By [Gin06] Proposition 5.3, [JL13] Theorem 1 and [CFK18] Theorem 5, the generalized Speh representations are representations of type (k,n)_D. ... However, this result is not true when D is not a field."

    For the quaternionic unitary extension, the inducing representation θ is required to be of type (k,n)_D with D a division quaternion algebra. The paper neither constructs θ nor proves the type property; Remark 2.13 delegates exactly this fact to [CFK18] Theorem 5, a paper with overlapping authorship. Remark 2.9 concedes that the usual nilpotent-orbit route to the multiplicity-one condition is false for non-field D, so the cited theorem is the only support for the quaternionic case. The Eulerian factorization and the stabilizer character χθ used in Theorem 5.3 and Section 5.2 are literally the defining content of Definition 2.8(1) and Definition 2.11(3), so the applicability of the main result to quaternionic unitary groups reduces to this self-cited existence claim.

full rationale

No equation-level circularity was found in the derivation of Theorem 5.3. Given an irreducible automorphic representation θ of type (k,n)_D, the paper proves the needed lemmas (2.15, 2.16, 2.17, 2.19, 6.5, 6.6, 6.7) directly from Definition 2.8/2.11, and the double-coset eliminations in Section 6 are linear algebra. No fitted parameter is renamed a prediction, and no identity is shown to be its own input. The flagged concern is the existence of the input class itself for quaternionic D: Remark 2.12 notes that for non-field D the local type condition does not follow from the global conditions, and Remark 2.13 routes the existence and type property of generalized Speh representations through [Gin06], [JL13], and [CFK18], the last of which shares authors with the present paper. The paper gives no verification that these references cover division quaternion D, and Remark 2.9 explicitly warns that the natural multiplicity-one route fails there. This makes the quaternionic extension depend on a self-cited, not independently checked, existence theorem. Since the central unfolding proof has substantial independent content, the appropriate finding is partial self-citation load-bearing rather than full circularity.

Assumptions & free parameters 0 free parameters · 3 assumptions · 0 invented entities

No free parameters appear in this pure mathematics paper. The main assumptions are standard representation-theoretic background plus the existence of type (k,n)_D representations; the latter is load-bearing for the quaternionic extension and is not proved in the paper.

assumptions (3)
  • domain assumption The inducing representation θ of GL_{kn,D}(A) is of type (k,n)_D, including local multiplicity one of degenerate Whittaker functionals.
    Invoked in Sections 4 and 5. Remark 2.9 states that multiplicity one in Definition 2.8(1) is not true when D is not a field; no construction is given for division quaternion D.
  • standard math Lemma 2.4: two maximal totally isotropic subspaces L,L' are P(Y)-equivalent iff the intersection dimensions match, and N(Y)-equivalent iff the successive quotients match.
    Proof is left to the reader in Section 2.3; the lemma underlies the orbit elimination in the unfolding proofs.
  • standard math Local representation theory tools: Bernstein-Zelevinsky geometric lemma, Bernstein's localization principle, and the theory of l-sheaves.
    Used in Section 7 to prove local multiplicity one. These are standard results in p-adic representation theory.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Twisted doubling integrals for classical groups." pith.science (2026). https://pith.science/paper/FWPAP5L7

@misc{pith2026190810298,
  author       = {Pith},
  title        = {Pith review of: Twisted doubling integrals for classical groups},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/FWPAP5L7}},
  note         = {Machine review of arXiv:1908.10298}
}
read the original abstract

We describe the twisted doubling integrals of Cai-Friedberg-Ginzburg-Kaplan in a conceptual way. This also extends the construction to the quaternionic unitary groups. We carry out the unfolding argument uniformly in this article. To do so, we define a family of degenerate Whittaker coefficients that are suitable in this setup and study some of their properties. We also prove certain related global and local results that use the same tools.

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

22 extracted references · 18 canonical work pages

  1. [1]

    I. N. Bernstein and A. V. Zelevinsky. Induced representations of reductive p -adic groups. I . Ann. Sci. \'Ecole Norm. Sup. (4) , 10(4):441--472, 1977

  2. [2]

    Bernstein

    Joseph N. Bernstein. P -invariant distributions on GL (N) and the classification of unitary representations of GL (N) (non- A rchimedean case). In Lie group representations, II ( C ollege P ark, M d., 1982/1983) , volume 1041 of Lecture Notes in Math. , pages 50--102. Springer, Berlin, 1984

  3. [3]

    I. N. Bern s te n and A. V. Zelevinski . Representations of the group GL(n,F), where F is a local non- A rchimedean field. Uspehi Mat. Nauk , 31(3(189)):5--70, 1976

  4. [4]

    Central extensions of reductive groups by K_2

    Jean-Luc Brylinski and Pierre Deligne. Central extensions of reductive groups by K_2 . Publ. Math. Inst. Hautes \'Etudes Sci. , (94):5--85, 2001

  5. [5]

    Automorphic forms and representations , volume 55 of Cambridge Studies in Advanced Mathematics

    Daniel Bump. Automorphic forms and representations , volume 55 of Cambridge Studies in Advanced Mathematics . Cambridge University Press, Cambridge, 1997

  6. [6]

    Doubling constructions and tensor product L -functions: the linear case

    Yuanqing Cai, Solomon Friedberg, David Ginzburg, and Eyal Kaplan. Doubling constructions and tensor product L -functions: the linear case. Invent. Math. , 217(3):985--1068, 2019

  7. [7]

    Doubling constructions: local and global theory, with an application to global functoriality for non-generic cuspidal representations

    Yuanqing Cai, Solomon Friedberg, and Eyal Kaplan. Doubling constructions: local and global theory, with an application to global functoriality for non-generic cuspidal representations. arXiv preprint arXiv:1802.02637 , 2018

  8. [8]

    Doubling zeta integrals and local factors for metaplectic groups

    Wee Teck Gan. Doubling zeta integrals and local factors for metaplectic groups. Nagoya Math. J. , 208:67--95, 2012

Show all 22 references
  1. [9]

    Certain conjectures relating unipotent orbits to automorphic representations

    David Ginzburg. Certain conjectures relating unipotent orbits to automorphic representations. Israel J. Math. , 151:323--355, 2006

  2. [10]

    Generalized and degenerate W hittaker models

    Raul Gomez, Dmitry Gourevitch, and Siddhartha Sahi. Generalized and degenerate W hittaker models. Compos. Math. , 153(2):223--256, 2017

  3. [11]

    Whittaker supports for representations of reductive groups

    Raul Gomez, Dmitry Gourevitch, and Siddhartha Sahi. Whittaker supports for representations of reductive groups. Ann. Inst. Fourier (Grenoble) , to appear

  4. [12]

    Generalized and degenerate W hittaker quotients and F ourier coefficients

    Dmitry Gourevitch and Siddhartha Sahi. Generalized and degenerate W hittaker quotients and F ourier coefficients. In Representations of reductive groups , volume 101 of Proc. Sympos. Pure Math. , pages 133--154. Amer. Math. Soc., Providence, RI, 2019

  5. [13]

    Kudla, and William J

    Michael Harris, Stephen S. Kudla, and William J. Sweet. Theta dichotomy for unitary groups. J. Amer. Math. Soc. , 9(4):941--1004, 1996

  6. [14]

    On F ourier coefficients of automorphic forms of GL (n)

    Dihua Jiang and Baiying Liu. On F ourier coefficients of automorphic forms of GL (n) . Int. Math. Res. Not. IMRN , (17):4029--4071, 2013

  7. [15]

    On the local factors of irreducible representations of quaternionic unitary groups

    Hirotaka Kakuhama. On the local factors of irreducible representations of quaternionic unitary groups. Manuscripta Math. , to appear

  8. [16]

    Doubling constructions and tensor product L -functions: coverings of the symplectic group

    Eyal Kaplan. Doubling constructions and tensor product L -functions: coverings of the symplectic group. arXiv preprint arXiv:1902.00880 , 2019

  9. [17]

    Martin L. Karel. Functional equations of W hittaker functions on p -adic groups. Amer. J. Math. , 101(6):1303--1325, 1979

  10. [18]

    Lapid and Stephen Rallis

    Erez M. Lapid and Stephen Rallis. On the local factors of representations of classical groups. In Automorphic representations, L -functions and applications: progress and prospects , volume 11 of Ohio State Univ. Math. Res. Inst. Publ. , pages 309--359. de Gruyter, Berlin, 2005

  11. [19]

    M glin and J.-L

    C. M glin and J.-L. Waldspurger. Mod\`eles de W hittaker d\'eg\'en\'er\'es pour des groupes p -adiques. Math. Z. , 196(3):427--452, 1987

  12. [20]

    Piatetski-Shapiro and S

    I. Piatetski-Shapiro and S. Rallis. L -functions for the classical groups , volume 1254 of Lecture Notes in Math. Springer-Verlag, New York, 1987

  13. [21]

    L-functions and theta correspondence for classical groups

    Shunsuke Yamana. L-functions and theta correspondence for classical groups. Invent. Math. , 196(3):651--732, 2014

  14. [22]

    A. V. Zelevinsky. Induced representations of reductive p -adic groups. II . O n irreducible representations of GL (n) . Ann. Sci. \'Ecole Norm. Sup. (4) , 13(2):165--210, 1980

Pith tools

Reviewed August 14, 2026 · model on record in the stance chip above.