Double logarithmic contribution to Higgs pair production in the high-energy limit

Tao Liu; Zhenghong Hu

REVIEW 3 major objections 5 minor 54 references

Double logarithmic contribution to Higgs pair production in the high-energy limit

T0 review · 3 major / 5 minor · reviewed 2026-08-04 · deepseek-v4-flash

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗

Pith's one-line read The authors compute the first three-loop abelian double-logarithmic QCD correction to Higgs pair production in the high-energy limit, fixing the rho^2 ln^6 rho coefficients of both box form factors as rational functions of s and t.

desk verdict First three-loop abelian double-log coefficient for gg→hh box form factors; benchmarked at one and two loops against [28], but the new three-loop result is stated without the derivation details needed to verify it, so plausible-conditional is the right verdict. read the letter →

arxiv 2509.06381 v1 pith:VEI5QLY2 submitted 2025-09-08 hep-ph

Zhenghong Hu , Tao Liu This is my paper · ORCID

classification hep-ph

keywords Higgspairproductiongluonfusiondoublelogarithmshigh-energylimitSudakovsoftquarkexchangethree-loopQCDboxformfactors

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

The paper addresses a gap in QCD corrections to Higgs pair production via gluon fusion: the leading double-logarithmic contributions that survive in the high-energy limit m_h^2 << m_t^2 << s, |t|. For the box-type Feynman diagrams, the authors compute the one- and two-loop leading logarithms with the Sudakov method, finding full agreement with the known analytic expansion, and then obtain the abelian (C_F^2) part at three loops for the first time. The new three-loop result is a finite rho^2 ln^6 rho term with rational coefficients in s and t, offered as a cross-check for future full calculations. The paper also shows that the double-logarithmic structure of the box diagrams is qualitatively richer than that of the triangle diagrams: new momentum configurations appear at every loop, so the simple soft-gluon-dressing picture fails and an all-order analysis will require a different method.

What carries the argument

The machinery is the Sudakov/eikonal analysis of soft quark lines. A soft loop momentum is written as l = q1 u + q2 v + l_perp with the constraint 1 > u v > rho, making the two propagators attached to it eikonal (1/u, 1/v), while the transverse component squared is identified with the quark mass, l_perp^2 -> m_t^2, which implements the power suppression in rho. The physical insight is that these double logarithms come from eikonal charge non-conservation with soft fermion exchange, not from gauge-boson Sudakov logs. This lets the authors classify configurations by which propagators are soft and by ordering constraints on the Sudakov variables (e.g., v2 > v1, u1 > v2), turning each region int

What would settle it

An independent three-loop expansion of the exact gg -> hh amplitude in the high-energy limit, extracting the coefficient of rho^2 ln^6 rho from F_box1 and F_box2, would settle the claim: agreement with Eq. (4.2) confirms the region enumeration, any deviation disproves it. Since the one- and two-loop checks against ref. [28] already passed, the decisive test is at three loops.

Watch

Extended reading notes

Core claim

In the limit m_h = 0, m_t^2 << s, |t|, the box form factors F_box1 and F_box2 of gg -> hh receive leading double-logarithmic corrections of the form rho^2 ln^{2L} rho at L loops, with rho = m_t^2/s, produced entirely by Sudakov-type soft quark exchange. After subtracting the conventional infrared-divergent Sudakov factor, the two-loop coefficients agree with the analytic expansion of ref. [28]. At three loops the authors compute the finite abelian corrections, finding F_box1^(2) = -C_F^2 (399 s^2 + 628 s t + 628 t^2)/(720 t (s+t)) rho^2 ln^6 rho and F_box2^(2) = -C_F^2 (40 s^2 + 43 s t + 43 t^2)/(180 t (s+t)) rho^2 ln^6 rho. A central structural finding is that each loop order introduces new

Load-bearing premise

The calculation assumes that every leading double logarithm comes from loop-momentum regions where the internal quark lines are soft in the Sudakov sense and all other momentum regions are power suppressed; if a non-soft region contributed at the same leading order, the coefficients of Eq. (4.2) would be incomplete.

Editorial extensions

If this is right

Eq. (4.2) supplies a concrete coefficient for any future full three-loop calculation of gg -> hh: matching it would validate the region analysis, and a mismatch would locate a missing soft configuration.
Through two loops, the Sudakov/eikonal method reproduces the known analytic leading-log coefficients of ref. [28], extending confidence in this method to four-point, two-Higgs amplitudes.
Triangle-type contributions remain tied to single-Higgs results [45-47], so the box diagrams are the only irreducible part requiring new all-order technology for this process.
The non-abelian three-loop leading logarithms remain uncomputed; the authors expect them to be tractable with the same tools, which would complete the three-loop picture.
Because new configurations appear at each loop order, the leading logarithms cannot be resummed by recursively dressing lower-loop diagrams; the paper's three-loop result is a step toward a genuinely all-order treatment.

Reading between the lines

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

A direct test of (4.2) could be made by numerically expanding the exact three-loop amplitude in the high-energy limit and extracting the rho^2 ln^6 rho coefficient; this would not require resolving the authors' region classification.
The exact cancellation of diagrams (l) and (m) before integration hints at a symmetry in the soft-phase-space measure; pinning it down could reduce the combinatorial count in higher loops.
The same soft-quark-exchange mechanism should generate analogous mass-suppressed double logarithms in other processes with two massive quark lines and multiple bosons, so the coefficient pattern here may transfer with color-factor replacements.
If the proliferation of new configurations continues, an effective-field-theory formulation that derives all regions from one soft function may outperform the diagrammatic enumeration, a possibility the paper's conclusion implicitly invites.

Share X Bluesky LinkedIn Reddit HN

Editorial analysis

A structured set of objections, weighed in public.

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

Desk Editor's Note

read the letter

Quick take: the useful new content is Eq. (4.2), the three-loop abelian leading-logarithmic coefficient for the box form factors in the high-energy limit. The one- and two-loop results agree with the independent benchmark [28], which gives real support to the Sudakov-region method as applied. No free parameters, no invented structures, and the paper is candid about what it does and does not do.

The weakness is exactly where the stress-test note points: the three-loop result is not auditable from the manuscript. Section 4 says "we refrain from discussing them in detail" for non-planar diagrams, and for Fig. 6(h) the paper says "we do not provide the detailed information... which should be easy to reconstruct." Table 1 gives only the final inequality constraints on Sudakov variables, with no per-configuration contributions, numerator reductions, or transverse integrals. The cancellations in Figs. 6(l,m) are asserted, not shown. So the 36 configurations for Fig. 6(h), the 24 diagrams in the Fig. 6(i) topology, and the claimed cancellations are all presented as summaries. A single missed region or a sign error in one of the many configurations would change Eq. (4.2) at the same order.

That does not mean the result is wrong. The one- and two-loop agreement with ref. [28] shows the framework can work, and the text explicitly says most three-loop regions are new, so that check cannot validate them. The load-bearing assumption is the completeness of the Sudakov/eikonal region enumeration — this is stated, but not proven. The scope is also narrower than the title might suggest: only the abelian, infrared-finite part at three loops; the non-abelian part and double-triangle diagrams are left for future work. That is a stated limitation, not a hidden flaw.

Citation pattern looks fine: the one- and two-loop check is against an independent group's calculation, and the self-citations are to earlier work developing the method. The paper is honest and the thinking is clear.

Who this is for: people working on high-energy gg→hh or on power-suppressed double logarithms in fermion-mediated processes. A serious referee should see it, but should insist on the full decomposition or an independent check before Eq. (4.2) is trusted as an established result. I would not desk-reject it, but I would treat the central new claim as conditional pending a fuller presentation.

Referee Report

3 major / 5 minor

Summary. This short paper addresses the leading double-logarithmic QCD corrections to gg→HH in the high-energy limit m_h^2 ≪ m_t^2 ≪ s,|t|. It uses Sudakov decomposition of soft quark momenta to isolate power-suppressed double logarithms, focusing on box-type Feynman diagrams. The paper reproduces the one-loop result (Eq. (2.9)) and the two-loop leading-logarithm result (Eq. (3.3)) for the box form factors, both in agreement with the independent calculation of ref. [28]. Its new result is Eq. (4.2), the three-loop abelian coefficient for F_box1 and F_box2, obtained by enumerating Sudakov regions in 13 three-loop abelian topologies. The paper also identifies new momentum configurations at two and three loops, notes the appearance of the first genuinely new three-loop configuration f3, and reports cancellations for Figs. 6(l,m).

Significance. If correct, Eq. (4.2) is the first three-loop abelian double-logarithmic result for gg→HH in the high-energy limit and provides a concrete target for future analytical or numerical cross-checks. The method is not new, but the paper extends it to a process whose double-logarithmic structure is substantially richer than that of single-Higgs production. The one- and two-loop agreement with the independent result of ref. [28] is a genuine nontrivial check of the method and is the strongest evidence in favor of the calculation. The main weakness is that the three-loop result is not auditable from the manuscript as written: the region enumeration, Sudakov decompositions, numerator reductions, transverse integrals, and per-configuration sums are largely omitted, and the text explicitly says that detailed discussions are omitted. This limits the usefulness of the claimed new result until the missing derivations are supplied.

major comments (3)

[Sec. 4, Eq. (4.2)] The central claim, Eq. (4.2), is not reproducible from the manuscript. The text explicitly says "we refrain from discussing them in detail" for the non-planar diagrams, and for Table 1 "we do not provide the detailed information on the decomposition of the soft loop momentum li, which should be easy to reconstruct." For Fig. 6(h) only the final count of 36 configurations and six constraint classes are given; for the topology of Fig. 6(i) there are "24 diagrams ... each one owns more than 20 configurations" with no individual results; the vanishing of Figs. 6(l,m) is stated as an accidental cancellation. A single missing or miscounted region, or an incorrect sign in one of the many regions, changes the rational functions in Eq. (4.2). The one- and two-loop agreement with ref. [28] validates the framework but does not validate this new three-loop coefficient. I request a complete derivatio
[Secs. 3 and 4, Table 1] The completeness of the manual enumeration of Sudakov regions is not established. At three loops the paper selects 13 typical diagrams and states that most configurations are new, but it does not describe a systematic rule that guarantees all leading-log regions have been found. In Table 1 for Fig. 6(h), the discarded candidates p1,3,7 and p1,2,6 are dismissed in one sentence each, and the constraints for the six non-vanishing classes are given without derivation. Since Eq. (4.2) depends directly on there being no missing region in any non-planar topology, this is a load-bearing gap. I ask for either a systematic region-finding procedure or a detailed account of why each discarded configuration is power suppressed.
[Sec. 4, Figs. 6(i), 6(l), 6(m)] The statements that Figs. 6(l,m) vanish due to "accidental cancellations between different configurations in each diagram" and that Fig. 6(i) has "more than 20 configurations to calculate" are not enough to reproduce the result. The latter is especially important because the text notes that different numerator choices lead to different constraints and overall coefficients. Without the explicit configuration sums, the cancellations and the quoted rational-function coefficients in Eq. (4.2) cannot be checked. Please list the non-vanishing contributions for each topology, including the numerator/constraint choices, and show the cancellations for Figs. 6(l,m) explicitly.

minor comments (5)

[Sec. 3] Typo: "Sukakov variables" should be "Sudakov variables."
[Table 1] The caption reads "T able 1" with a space, and "Table. 1" appears in the text; please fix the formatting.
[Fig. 7] The sentence "the unshown f1,...,6 have the same distributions as e1,...,6" is terse; please clarify how the f configurations are obtained from the e configurations by external-leg permutations and state the multiplicities explicitly.
[Eq. (2.9)] The one-loop result F_box2=0 is stated after summing; a brief explanation of why the two Lorentz structures receive different leading-log contributions would improve readability.
[Sec. 4, Eq. (4.2)] The superscript notation in F_box1^(2) may confuse readers because Eq. (2.4) uses the same superscript for the expansion in α_s. Please state explicitly that the superscript (2) denotes three-loop order relative to the leading amplitude (i.e., the α_s^2 term in the form-factor expansion of Eq. (2.4)).

Circularity Check

0 steps flagged · score 0.0 of 10

No load-bearing circularity: the three-loop abelian box coefficient is computed from Sudakov regions and benchmarked against an external result, not imported from inputs.

full rationale

The central result, Eq. (4.2), is not obtained by fitting, by renaming a known result, or by importing a uniqueness/ansatz theorem from the authors' prior work. The one-loop result (2.9) and two-loop result (3.3) are compared with the independent analytical results of ref. [28], so the Sudakov-region method is externally benchmarked at the orders where it is used. The citations to the authors' earlier papers [45-47] concern the triangle-diagram contributions and the IR-subtraction prescription, not the box-diagram leading logarithms that produce the new three-loop coefficient; even if those citations are self-referential, they do not fix the rational functions in Eq. (4.2). The paper's own caveat that detailed Sudakov decompositions are omitted (Sec. 4, 'For simplicity, we do not provide the detailed information on the decomposition of the soft loop momentum li, which should be easy to reconstruct') is a reproducibility/auditability limitation, not a circularity: an omitted region would change the answer, but the paper does not define the input as the output, and no fitted parameter is renamed as a prediction. Accordingly, no circular step satisfying the definitional, fitted-input, self-citation-load-bearing, uniqueness-imported, ansatz-smuggled, or renaming patterns is present.

Assumptions & free parameters 0 free parameters · 4 assumptions · 0 invented entities

The calculation is a fixed-order perturbative QCD analysis with no fitted parameters and no newly postulated physical entities. It relies on the standard QCD framework, the high-energy kinematic limit, and the Sudakov/eikonal approximation inherited from the authors' prior work and related literature.

assumptions (4)

standard math Standard QCD Feynman rules and SU(N_c) color algebra with CF, CA, TF.
Used throughout to write the amplitude decomposition and color factors in Eqs. (2.2)-(2.5), (3.3), and (4.2).
domain assumption High-energy kinematic limit m_h^2 << m_t^2 << s, |t|, with m_h set to zero from the start and log(m_t^2/|t|) approximated by log(m_t^2/s), rho = m_t^2/s.
Defines the regime of the calculation; stated in the introduction and used in the Sudakov analysis and in the final logarithms.
domain assumption Sudakov/eikonal method: leading double logarithms arise only from momentum configurations where internal quark propagators are soft, satisfying inequalities on the Sudakov variables ui, vi; other momentum regions are power suppressed.
This is the load-bearing premise behind the selection of non-vanishing configurations in Sections 3 and 4, e.g., Eqs. (3.1)-(3.2) and Table 1.
domain assumption IR divergences of the two-loop box diagrams factorize as in light-quark mediated single Higgs production and can be subtracted in factorized form before comparison with ref. [28].
Invoked in Section 3: 'As done in refs. [45-47] IR divergences are subtracted in the factorized form of simple integrals'. This is needed for the one- and two-loop agreement and for defining the finite remainder.

how reviews work

0 comments

Cite this review

Pith. "Pith review of Double logarithmic contribution to Higgs pair production in the high-energy limit." pith.science (2026). https://pith.science/paper/VEI5QLY2

@misc{pith2026250906381,
  author       = {Pith},
  title        = {Pith review of: Double logarithmic contribution to Higgs pair production in the high-energy limit},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/VEI5QLY2}},
  note         = {Machine review of arXiv:2509.06381}
}

read the original abstract

We study the leading logarithmic QCD corrections to Higgs pair production in the high-energy limit, which originate from soft quark exchange in the Feynman diagram and are thus suppressed by the quark mass. It is found that triangle Feynman diagrams can be analyzed in the same way as single Higgs production mediated by bottom quarks [45-47], while for diagrams of box type we cannot get an all-order result due to the many more double logarithmic structures they own. In this paper we calculate the abelian corrections at three loops for the first time after obtaining the corresponding one- and two-loop leading logarithms, which are in total agreement with the analytical results in the literature.

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

54 extracted references · 52 canonical work pages

[28]

Davies, G

J. Davies, G. Mishima, M. Steinhauser and D. Wellmann, J HEP 01 (2019), 176

work page 2019
[1]

de Florian et al

D. de Florian et al. [LHC Higgs Cross Section Working Group], CERN Yellow Rep. Mo nogr. 2 (2017), 1-869

work page 2017
[2]

E. W. N. Glover and J. J. van der Bij, Nucl. Phys. B 309 (1988), 282-294

work page 1988
[3]

Plehn, M

T. Plehn, M. Spira and P. M. Zerwas, Nucl. Phys. B 479 (1996), 46-64

work page 1996
[4]

Borowka, N

S. Borowka, N. Greiner, G. Heinrich, S. P. Jones, M. Kerne r, J. Schlenk, U. Schubert and T. Zirke, Phys. Rev. Lett. 117 (2016) no.1, 012001

work page 2016
[5]

Borowka, N

S. Borowka, N. Greiner, G. Heinrich, S. P. Jones, M. Kerne r, J. Schlenk and T. Zirke, JHEP 10 (2016), 107

work page 2016
[6]

Baglio, F

J. Baglio, F. Campanario, S. Glaus, M. Mühlleitner, M. Sp ira and J. Streicher, Eur. Phys. J. C 79 (2019) no.6, 459

work page 2019
[7]

Davies, G

J. Davies, G. Heinrich, S. P. Jones, M. Kerner, G. Mishima , M. Steinhauser and D. Wellmann, JHEP 11 (2019), 024

work page 2019

Show all 54 references

[8]

J. M. Campbell, G. De Laurentis and R. K. Ellis, JHEP 10 (2024), 230

2024
[9]

Heinrich, S

G. Heinrich, S. P. Jones, M. Kerner, G. Luisoni and E. Vryo nidou, JHEP 08 (2017), 088

2017
[10]

Jones and S

S. Jones and S. Kuttimalai, JHEP 02 (2018), 176

2018
[11]

Heinrich, S

G. Heinrich, S. P. Jones, M. Kerner, G. Luisoni and L. Scy boz, JHEP 06 (2019), 066

2019
[12]

Bagnaschi, G

E. Bagnaschi, G. Degrassi and R. Gröber, Eur. Phys. J. C 83 (2023) no.11, 1054

2023
[13]

De Florian and J

D. De Florian and J. Mazzitelli, JHEP 08 (2018), 156

2018
[14]

de Florian, M

D. de Florian, M. Grazzini, C. Hanga, S. Kallweit, J. M. L indert, P. Maierhöfer, J. Mazzitelli and D. Rathlev, JHEP 09 (2016), 151

2016
[15]

Grigo, J

J. Grigo, J. Hoﬀ and M. Steinhauser, Nucl. Phys. B 900 (2015), 412-430

2015
[16]

Davies, F

J. Davies, F. Herren, G. Mishima and M. Steinhauser, JHE P 01 (2022), 049

2022
[17]

Davies, K

J. Davies, K. Schönwald and M. Steinhauser, Phys. Lett. B 845 (2023), 138146

2023
[18]

Davies, K

J. Davies, K. Schönwald, M. Steinhauser and M. Vitti, JH EP 08 (2024), 096

2024
[19]

Grazzini, G

M. Grazzini, G. Heinrich, S. Jones, S. Kallweit, M. Kern er, J. M. Lindert and J. Mazzitelli, JHEP 05 (2018), 059

2018
[20]

L. B. Chen, H. T. Li, H. S. Shao and J. Wang, Phys. Lett. B 803 (2020), 135292

2020
[21]

L. B. Chen, H. T. Li, H. S. Shao and J. Wang, JHEP 03 (2020), 072

2020
[22]

A H and H

A. A H and H. S. Shao, JHEP 02 (2023), 067 – 13 –

2023
[23]

H. Y. Bi, L. H. Huang, R. J. Huang, Y. Q. Ma and H. M. Yu, Phys . Rev. Lett. 132 (2024) no.23, 231802

2024
[24]

Heinrich, S

G. Heinrich, S. Jones, M. Kerner, T. Stone and A. Vestner , JHEP 11 (2024), 040

2024
[25]

Gröber, A

R. Gröber, A. Maier and T. Rauh, JHEP 03 (2018), 020

2018
[26]

Bonciani, G

R. Bonciani, G. Degrassi, P. P. Giardino and R. Gröber, P hys. Rev. Lett. 121 (2018) no.16, 162003

2018
[27]

Davies, G

J. Davies, G. Mishima, M. Steinhauser and D. Wellmann, J HEP 03 (2018), 048

2018
[29]

Davies, G

J. Davies, G. Mishima, K. Schönwald, M. Steinhauser and H. Zhang, JHEP 08 (2022), 259

2022
[30]

Davies, K

J. Davies, K. Schönwald, M. Steinhauser and H. Zhang, JH EP 10 (2023), 033

2023
[31]

Davies, K

J. Davies, K. Schönwald, M. Steinhauser and H. Zhang, JH EP 04 (2025), 193

2025
[32]

Bonetti, P

M. Bonetti, P. Rendler and W. J. Torres Bobadilla, JHEP 07 (2025), 024

2025
[33]

Z. Hu, T. Liu and J. M. Yang, [arXiv:2503.10051 [hep-ph] ]

arXiv
[34]

M. I. Kotsky and O. I. Yakovlev, Phys. Lett. B 418, 335 (1998)

1998
[35]

A. A. Penin, Phys. Lett. B 745, 69 (2015)

2015
[36]

Melnikov and A

K. Melnikov and A. Penin, JHEP 05 (2016), 172

2016
[37]

A. A. Penin and N. Zerf, Phys. Lett. B 760, 816 (2016)

2016
[38]

T. Liu, A. A. Penin and N. Zerf, Phys. Lett. B 771, 492 (2017)

2017
[39]

V. V. Sudakov, Sov. Phys. JETP 3, 65 (1956) [Zh. Eksp. Teor. Fiz. 30, 87 (1956)]

1956
[40]

Frenkel and J

J. Frenkel and J. C. Taylor, Nucl. Phys. B 116, 185 (1976)

1976
[41]

A. H. Mueller, Phys. Rev. D 20, 2037 (1979)
[42]

J. C. Collins, Phys. Rev. D 22, 1478 (1980)

1980
[43]

Sen, Phys

A. Sen, Phys. Rev. D 24, 3281 (1981)

1981
[44]

G. F. Sterman, Nucl. Phys. B 281, 310 (1987)

1987
[45]

Liu and A

T. Liu and A. A. Penin, Phys. Rev. Lett. 119 (2017) no.26, 262001

2017
[46]

Liu and A

T. Liu and A. Penin, JHEP 11 (2018), 158

2018
[47]

T. Liu, S. Modi and A. A. Penin, JHEP 02 (2022), 170

2022
[48]

T. Liu, A. A. Penin and A. Rehman, JHEP 04 (2024), 031

2024
[49]

Anastasiou and A

C. Anastasiou and A. Penin, JHEP 07 (2020), 195

2020
[50]

Z. L. Liu, B. Mecaj, M. Neubert and X. Wang, JHEP 01 (2021), 077

2021
[51]

Z. L. Liu, M. Neubert, M. Schnubel and X. Wang, JHEP 06 (2023), 183

2023
[52]

J. Y. Hou, J. Wang and D. J. Zhang, JHEP 06 (2025), 164

2025
[53]

Jaskiewicz, S

S. Jaskiewicz, S. Jones, R. Szafron and Y. Ulrich, [arXi v:2501.00587 [hep-ph]]

arXiv
[54]

Binosi and L

D. Binosi and L. Theussl, Comput. Phys. Commun. 161 (2004), 76-86 – 14 –

2004

Pith tools

Reviewed August 4, 2026 · model on record in the stance chip above.

[1] [28]

Davies, G

J. Davies, G. Mishima, M. Steinhauser and D. Wellmann, J HEP 01 (2019), 176

work page 2019

[2] [1]

de Florian et al

D. de Florian et al. [LHC Higgs Cross Section Working Group], CERN Yellow Rep. Mo nogr. 2 (2017), 1-869

work page 2017

[3] [2]

E. W. N. Glover and J. J. van der Bij, Nucl. Phys. B 309 (1988), 282-294

work page 1988

[4] [3]

Plehn, M

T. Plehn, M. Spira and P. M. Zerwas, Nucl. Phys. B 479 (1996), 46-64

work page 1996

[5] [4]

Borowka, N

S. Borowka, N. Greiner, G. Heinrich, S. P. Jones, M. Kerne r, J. Schlenk, U. Schubert and T. Zirke, Phys. Rev. Lett. 117 (2016) no.1, 012001

work page 2016

[6] [5]

Borowka, N

S. Borowka, N. Greiner, G. Heinrich, S. P. Jones, M. Kerne r, J. Schlenk and T. Zirke, JHEP 10 (2016), 107

work page 2016

[7] [6]

Baglio, F

J. Baglio, F. Campanario, S. Glaus, M. Mühlleitner, M. Sp ira and J. Streicher, Eur. Phys. J. C 79 (2019) no.6, 459

work page 2019

[8] [7]

Davies, G

J. Davies, G. Heinrich, S. P. Jones, M. Kerner, G. Mishima , M. Steinhauser and D. Wellmann, JHEP 11 (2019), 024

work page 2019