Recognition: no theorem link
Remarks on the distribution of Dirichlet L-functions along cosets
Pith reviewed 2026-05-11 03:04 UTC · model grok-4.3
The pith
A modified CFKRS recipe accounts for root number dependence on coefficients to predict the secondary main term in second moments of Dirichlet L-functions along cosets.
A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.
Core claim
The original CFKRS recipe gives the incorrect answer for the second moment of Dirichlet L-functions along cosets because the root number is not always independent of the Dirichlet series coefficients along certain cosets. The proposed modification takes this feature into account and correctly predicts the secondary main term. Additional reformulations of classical methods in coset terms are used to bound hybrid moments and reinterpret van der Corput's bound.
What carries the argument
Modified CFKRS recipe that incorporates the dependence between root numbers and Dirichlet coefficients along cosets
If this is right
- The second moment asymptotic along cosets includes a specific secondary main term correctly predicted by the modified recipe.
- Heath-Brown's q-analog of van der Corput's shifting method reformulated in coset terms gives an upper bound for a hybrid second moment.
- The classical van der Corput bound is reinterpreted as an amplified second moment of a trigonometric polynomial.
- A collection of further problems on the distribution of Dirichlet L-functions along cosets admit similar treatments.
Where Pith is reading between the lines
- Similar root-number dependencies may affect moment predictions in other arithmetic families, such as quadratic twists, where independence assumptions are commonly used.
- The modified recipe could be tested directly by computing moments numerically over small moduli and comparing the observed secondary coefficient to the prediction.
- This adjustment suggests that random-matrix models for L-functions require case-by-case verification of coefficient-root-number independence when restricting to arithmetic progressions or cosets.
Load-bearing premise
The dependence between the root number and the Dirichlet series coefficients is the sole reason the original recipe fails for cosets, and the proposed modification fully captures the secondary main term without additional unaccounted contributions.
What would settle it
Numerical computation of the second moment along a fixed coset for a long range of conductors, checking whether the coefficient of the secondary term matches the modified recipe's prediction exactly.
read the original abstract
In a previous work with B. Garcia, the author considered the asymptotic for the second moment of Dirichlet $L$-functions along cosets, and exhibited a surprising secondary main term that is not predicted by the recipe of Conrey, Farmer, Keating, Rubinstein, and Snaith. In this paper, we re-examine this problem and propose a modified recipe that correctly predicts this secondary main term. The original recipe gives the incorrect answer for this family because the root number is not always independent of the Dirichlet series coefficients along certain cosets, and our proposed fix simply takes this feature into account. In addition, we consider a handful of other problems related to Dirichlet $L$-functions along cosets. One goal is to reformulate Heath-Brown's $q$-analog of van der Corput's shifting method in terms of cosets, which leads to an upper bound on a hybrid second moment. We also revisit the classical van der Corput bound and view it (in more modern terms) as an amplified second moment of a trigonometric polynomial.
Editorial analysis
A structured set of objections, weighed in public.
Referee Report
Summary. The paper proposes a modified version of the Conrey-Farmer-Keating-Rubinstein-Snaith (CFKRS) recipe for the second moment of Dirichlet L-functions restricted to cosets of characters modulo q. The modification accounts for the observed dependence between the root number and the Dirichlet coefficients along these cosets, which is claimed to explain and correctly predict the secondary main term found in the author's prior joint work with B. Garcia. The paper also reformulates Heath-Brown's q-analog of van der Corput's shifting method in coset language to obtain an upper bound for a hybrid second moment and reinterprets the classical van der Corput bound as an amplified second moment of a trigonometric polynomial.
Significance. If the modified recipe is shown to be complete and free of additional unaccounted terms, the work would usefully illustrate how standard random-matrix heuristics must be adjusted when arithmetic families impose linear constraints on the coefficients, with potential applicability to other restricted families. The coset reformulations of classical bounds provide a modern perspective that may aid in unifying amplification techniques across different contexts.
major comments (3)
- [§2] §2 (modified recipe): The central claim that adjusting solely for root-number/coefficient dependence exhausts the secondary main term is load-bearing, yet the derivation does not explicitly expand the approximate functional equation sum over the coset to confirm that correlations among Euler factors at distinct primes (induced by the residue-class constraint) contribute no further secondary terms. A complete expansion or explicit verification that such contributions vanish or are absorbed would be required.
- [§3] §3 (prior asymptotics re-examination): The re-derivation of the asymptotics from the earlier Garcia collaboration relies on the modified recipe matching the observed secondary term, but without a direct comparison of error terms or numerical verification for small q, it remains unclear whether the agreement is exact or merely asymptotic to leading order.
- [§4] §4 (Heath-Brown reformulation): The upper bound for the hybrid second moment is obtained by expressing the sum over cosets, but the uniformity of the error term with respect to the conductor and the coset representative is not stated explicitly; this affects whether the bound improves on or matches the original Heath-Brown result in the stated range.
minor comments (2)
- The introduction could include a brief diagram or explicit definition of the coset notation (e.g., the residue class condition) to make the paper more self-contained for readers who have not consulted the prior Garcia work.
- [§5] In the van der Corput reinterpretation, the transition from the trigonometric polynomial to the L-function moment could be accompanied by a short remark on how the amplification parameter is chosen, for clarity.
Simulated Author's Rebuttal
We thank the referee for their careful reading and constructive comments. We address each major comment below and indicate the revisions that will be incorporated.
read point-by-point responses
-
Referee: [§2] §2 (modified recipe): The central claim that adjusting solely for root-number/coefficient dependence exhausts the secondary main term is load-bearing, yet the derivation does not explicitly expand the approximate functional equation sum over the coset to confirm that correlations among Euler factors at distinct primes (induced by the residue-class constraint) contribute no further secondary terms. A complete expansion or explicit verification that such contributions vanish or are absorbed would be required.
Authors: We agree that an explicit verification strengthens the argument. In the revised manuscript we will add a detailed expansion of the approximate functional equation summed over the coset. This expansion shows that correlations among Euler factors at distinct primes, induced by the residue-class constraint, average to zero or are absorbed into the error term by character orthogonality and do not produce additional secondary main terms at the order under consideration. revision: yes
-
Referee: [§3] §3 (prior asymptotics re-examination): The re-derivation of the asymptotics from the earlier Garcia collaboration relies on the modified recipe matching the observed secondary term, but without a direct comparison of error terms or numerical verification for small q, it remains unclear whether the agreement is exact or merely asymptotic to leading order.
Authors: The modified recipe is constructed to reproduce the secondary main term exactly as observed in the prior work. The error terms remain those established in the Garcia collaboration and are of strictly lower order. In the revision we will insert an explicit comparison of these error terms and add a short numerical check for small q to illustrate that the agreement holds beyond the leading terms. revision: yes
-
Referee: [§4] §4 (Heath-Brown reformulation): The upper bound for the hybrid second moment is obtained by expressing the sum over cosets, but the uniformity of the error term with respect to the conductor and the coset representative is not stated explicitly; this affects whether the bound improves on or matches the original Heath-Brown result in the stated range.
Authors: Our coset reformulation preserves the uniformity present in Heath-Brown’s original argument. In the revised version we will explicitly record the uniformity of the error term in both the conductor and the coset representative, confirming that the resulting bound matches the original Heath-Brown result throughout the stated range. revision: yes
Circularity Check
No circularity: modification derived from identified independence failure, not fitted to output
full rationale
The paper cites prior joint work only to establish the existence of an unexpected secondary term in the coset second moment. It then identifies the root-number/coefficient dependence as the cause and states that the proposed recipe modification simply incorporates this dependence. No equation in the abstract or described chain reduces the secondary term to a fitted parameter or to a self-citation; the fix is presented as a direct accounting for a structural feature of the family. The derivation therefore remains independent of the target asymptotic and does not meet any of the enumerated circularity patterns.
Axiom & Free-Parameter Ledger
axioms (2)
- standard math Standard analytic continuation and functional equation properties of Dirichlet L-functions hold along cosets.
- domain assumption The CFKRS recipe framework applies to families of L-functions with the usual independence assumptions.
Reference graph
Works this paper leans on
-
[1]
Odoni, R. , TITLE =. Bull. London Math. Soc. , FJOURNAL =. 1973 , PAGES =. doi:10.1112/blms/5.3.325 , URL =
-
[2]
Petrow, Ian and Young, Matthew P. , TITLE =. Duke Math. J. , FJOURNAL =. 2023 , NUMBER =. doi:10.1215/00127094-2022-0069 , URL =
-
[3]
Heath-Brown, D. R. , TITLE =. Invent. Math. , FJOURNAL =. 1978 , NUMBER =. doi:10.1007/BF01578069 , URL =
-
[4]
Garcia, Bradford and Young, Matthew P. , TITLE =. Forum Math. Sigma , FJOURNAL =. 2025 , PAGES =. doi:10.1017/fms.2025.44 , URL =
-
[5]
Conrey, J. B. and Farmer, D. W. and Keating, J. P. and Rubinstein, M. O. and Snaith, N. C. , TITLE =. Proc. London Math. Soc. (3) , FJOURNAL =. 2005 , NUMBER =. doi:10.1112/S0024611504015175 , URL =
-
[6]
Kaczorowski, Jerzy and Molteni, Giuseppe and Perelli, Alberto , TITLE =. Comment. Math. Helv. , FJOURNAL =. 2010 , NUMBER =. doi:10.4171/CMH/202 , URL =
-
[7]
Molteni, G. , TITLE =. Acta Arith. , FJOURNAL =. 2010 , NUMBER =. doi:10.4064/aa145-1-3 , URL =
-
[8]
Marc Munsch and Igor E. Shparlinski , journal=. Moments and non-vanishing of
-
[9]
Large values of $L(\sigma,\chi)$ for subgroups of characters
Pranendu Darbar and Bryce Kerr and Marc Munsch and Igor Shparlinski , eprint=. Large values of. arXiv preprint arXiv:2604.02960 , year =
work page internal anchor Pith review Pith/arXiv arXiv
-
[10]
Khan, Rizwanur and Mili\'. Non-vanishing of. Int. Math. Res. Not. IMRN , FJOURNAL =. 2016 , NUMBER =. doi:10.1093/imrn/rnv320 , URL =
-
[11]
Iwaniec, Henryk and Kowalski, Emmanuel , TITLE =. 2004 , PAGES =. doi:10.1090/coll/053 , URL =
-
[12]
Heath-Brown, D. R. , TITLE =. Comment. Math. Helv. , FJOURNAL =. 1981 , NUMBER =. doi:10.1007/BF02566206 , URL =
-
[13]
Graham, S. W. and Kolesnik, G. , TITLE =. 1991 , PAGES =. doi:10.1017/CBO9780511661976 , URL =
-
[14]
Blomer, Valentin and Khan, Rizwanur and Young, Matthew , TITLE =. Duke Math. J. , FJOURNAL =. 2013 , NUMBER =. doi:10.1215/00127094-2380967 , URL =
-
[15]
Mili\'. Twelfth moment of. Ann. Sc. Norm. Super. Pisa Cl. Sci. (5) , FJOURNAL =. 2021 , NUMBER =
work page 2021
-
[16]
Nunes, Ramon M. , TITLE =. Int. Math. Res. Not. IMRN , FJOURNAL =. 2021 , NUMBER =. doi:10.1093/imrn/rnz089 , URL =
-
[17]
Heath-Brown, D. R. , TITLE =. Quart. J. Math. Oxford Ser. (2) , FJOURNAL =. 1978 , NUMBER =. doi:10.1093/qmath/29.4.443 , URL =
- [18]
discussion (0)
Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.