Field-theoretical approach to neutral pion contribution to muon $g-2$

Masashi Hayakawa

REVIEW 3 major objections 5 minor 16 references

Field-theoretical approach to neutral pion contribution to muon $g-2$

T0 review · 3 major / 5 minor · reviewed 2026-08-10 · deepseek-v4-flash

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

Pith's one-line read The neutral pion contributes to muon g-2 only through its on-shell pole, with no principal-value part.

desk verdict A promising critique of the standard pion-pole formula whose central conclusion—no principal-value term—is likely an artifact of a nonstandard regularization of the time integral. read the letter →

arxiv 2501.16547 v2 pith:T54X2EQY submitted 2025-01-27 hep-ph hep-lat

Masashi Hayakawa This is my paper · ORCID

classification hep-phhep-lat

keywords muong-2hadroniclight-by-lightscatteringneutralpionpoletransitionformfactorprincipalvalueQCDintermediatestatestime-orderedproductWickrotation

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 tries to settle what is meant by the neutral pion contribution to the muon anomalous magnetic moment $g-2$. Its proposal is to define that contribution directly as the single-pion intermediate state in the QCD vacuum expectation value of the time-ordered product of four electromagnetic currents, rather than to take the commonly used pion-pole formula as given. Starting from that definition, the paper derives Eq. (30): the neutral pion contribution is a sum of three delta-function terms, one in each of the $s$, $u$, and $t$ channels, with no principal-value part from the pion propagator. The paper concludes that the standard formulas used in model and lattice estimates of the pion-pole contribution contain an off-shell artifact, and that the accepted numerical pipelines for this piece of $a_\mu$ need to be rebuilt from the field-theoretic definition.

What carries the argument

The central machinery is the insertion of a complete set of QCD states into $\langle0|\mathcal{T}J_{\nu_1}(z_1)J_{\nu_2}(z_2)J_{\nu_3}(z_3)J_\mu(0)|0\rangle$, organized by $\theta$-functions so that a single pion intermediate state is singled out in each of the six pairings. The decisive step is the evaluation of the restricted time integral $I(\omega)$ of Eq. (21), which is ambiguous because $\int_0^\infty dt\,e^{i\omega t}$ has no unique value; the paper sets this integral proportional to $\delta(\omega)$ with an undetermined constant $\beta$, which removes the principal-value part $i\,\mathrm{PV}(1/\omega)$ from the final amplitude. Time-reversal symmetry is used to convert the pion matrix element with two currents into the standard on-shell transition form factor $F$, and the delta functions enforce energy-momentum conservation so that the pion is on its mass shell. The result is Eq. (30), the delta-function-only pole contribution, with $\beta$ fixed to $1$ by comparison with the Wess-Zumino-Witten low-energy effective theory.

What would settle it

Re-evaluate the integral $I(\omega)$ in Eq. (21) using the standard identity $\int_0^\infty dt\,e^{i\omega t}=\pi\delta(\omega)+i\,\mathrm{PV}(1/\omega)$; if a principal-value contribution reappears in the analogue of Eq. (30), the central claim collapses under the usual distributional convention.

Watch

Extended reading notes

Core claim

Singling out the neutral pion intermediate state in the time-ordered product of four quark electromagnetic currents, and using time-reversal symmetry to express the pion matrix elements through the physical on-shell transition form factor $F$, the paper obtains Eq. (30). That expression has three terms, one for each of the $s$, $u$, and $t$ channels, each proportional to $\delta((k_i+k_j)^2 - m_\pi^2)$ times a product of two transition form factors. No term corresponds to the principal value $\mathrm{PV}(1/(s-m_\pi^2))$ that would come from the real part of the pion propagator $1/((k_1+k_2)^2-m_\pi^2+i\epsilon)$. The author concludes that the standard formula (1) is not the correct field-theoretic neutral pion contribution: only its delta-function pole part survives, so the conventional pion-pole integrals of Refs. [15] and [16] are inapplicable because they treat the two loop time components as independent integration variables. The overall normalization $\beta$ is left undetermined by the field-theoretic derivation and is fixed to $\beta=1$ by matching the low-energy effective theory.

Load-bearing premise

The argument's load-bearing premise is a chosen rule for an ambiguous integral: the paper treats $\int_0^\infty dt\,e^{i\omega t}$ as proportional to $\delta(\omega)$ and discards the $i\,\mathrm{PV}(1/\omega)$ part that the standard distributional identity would give, and this choice is exactly what removes the principal-value terms from the final answer.

Editorial extensions

If this is right

The pion-pole contribution to $a_\mu$ must be evaluated from Eq. (30), not from the standard formulas of Refs. [15,16].
The only surviving pion propagator part is the imaginary delta-function piece, so the principal-value terms in Eq. (1) are an artifact of the off-shell extrapolation.
Because the delta function ties the two loop time components together, only one Wick rotation is available, and the contribution to $g-2$ comes from the imaginary part of the hadronic light-by-light amplitude.
Matching the low-energy effective theory fixes the overall normalization to $\beta=1$, so the on-shell pole term exactly reproduces the delta-function part of the conventional formula in the Wess-Zumino-Witten limit.

Reading between the lines

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

If the delta-function-only result is correct, the numerical pion-pole part of $a_\mu$ could shift once recomputed with Eq. (30), because the discarded principal-value terms previously entered through off-shell form-factor extrapolations.
The same complete-state insertion method could be applied to the $\eta$ and $\eta'$ poles, where an analogous off-shell ambiguity likely appears.
The alternative insertion in Eq. (35), which has no single neutral pion term, suggests a testable route to an independent estimate of hadronic light-by-light through $\pi^+\pi^-\gamma^*$ matrix elements.

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

Let me give you the short version: this paper does not establish the claim that the neutral pion contribution to hadronic light-by-light contains only the on-shell pole and no principal-value piece. The step that removes the PV terms is the treatment of the half-line Fourier integral in Eqs. (21)–(24), where the author defines ∫₀^∞ dt e^{iωt} to be a delta function. The standard distributional identity gives πδ(ω) + i PV(1/ω); applying it to I(ω) leaves PV terms, and there is no argument that they cancel after the remaining integrals. So the main result (30) is, at least partly, an input rather than an output.

That said, the paper is worth reading. Hayakawa is a serious person in this field, and he has put his finger on a genuine tension: the standard pion-pole formula (1) uses off-shell pions and a propagator with a PV part, while the transition form factor is defined only for on-shell pions. His explicit derivation from the four-current correlator, using time-reversal to express the pion matrix element, is careful and new. It cleanly reproduces the on-shell (delta) part of Eq. (1) up to the unknown normalization β, and it makes the point that the Gegenbauer-based formulas assume independent loop momenta, which contradicts the delta-function constraint. That challenge to Knecht–Nyffeler and the dispersive treatment is worth taking seriously.

The soft spots are as follows. The ambiguous integral in Eq. (23) is the load-bearing joint. The author acknowledges the ambiguity and then chooses a convention; that choice is not derived from QCD, and it is the choice that eliminates the PV terms. Also, β is determined only by matching to WZW, which is itself the low-energy theory that gives the standard formula; so the derivation does not independently fix the normalization. The paper contains no numerical estimate, so the magnitude of the claimed change is unknown. The references are the standard ones, including relevant self-citations, and that is not a problem here.

Who should read it: anyone working on the hadronic light-by-light contribution to g-2, especially those using the pion-pole formulas. It is a cautionary analysis, not a replacement result. If I were the editor, I would send it to a referee rather than desk-reject. The question—is the standard pion-pole formula for the neutral pion contribution actually justified?—is important for the g-2 program, and a careful referee might either find a physical reason for the chosen regularization or a clean counterexample. But as it stands, the central claim does not survive standard distributional reasoning. I would not cite it yet, though I might mention it as a cautionary note in a review.

Referee Report

3 major / 5 minor

Summary. The paper claims to derive, from the QCD four-current correlator, the neutral pion intermediate-state contribution to hadronic light-by-light scattering, and concludes that this contribution consists only of delta-function (on-shell pion pole) terms, with no principal-value terms. The derivation inserts a complete set of QCD states into the time-ordered product, uses time-reversal to express pion matrix elements in terms of the transition form factor, and obtains Eq. (30), which contains only δ((k_i+k_j)^2−m_π^2) terms. The overall normalization is left as an unfixed parameter β=2θ(0) and is later set to 1 by matching to the Wess-Zumino-Witten low-energy effective theory. The paper further argues that this result invalidates the standard pion-pole formulas of Refs. [15,16] for the muon g−2, because the delta functions impose kinematic constraints that make the usual Wick-rotation treatment inapplicable.

Significance. If the central claim were established, it would challenge the standard dispersive and pion-pole treatments of the hadronic light-by-light contribution to the muon g−2, a quantity of active experimental and theoretical interest. The paper is honest about its limitations: the normalization β is not derived from QCD and is fixed only by matching to the low-energy effective theory, and the author explicitly acknowledges that other insertions of a complete set of states lead to different decompositions. The derivation is presented in detail and the problematic step is identified openly. However, the main claim rests on a nonstandard treatment of an ambiguous oscillatory integral, and the paper does not provide a derivation of the required distributional identity from QCD or any physical principle. The significance of the intended result is high, but the correctness is not established by the present argument.

major comments (3)

[Sec. 3, Eqs. (21)–(25)] The central result Eq. (30) depends on the definition of the ambiguous integral I(ω) in Eq. (21). In Eqs. (23)–(24), the author defines ∫_0^∞ dt e^{iωt} to yield a delta function only, discarding the standard principal-value part. The standard distributional identity is πδ(ω) + i PV(1/ω) (up to convention); applying it to Eq. (22) produces products of delta functions and principal-value terms, including terms of the form PV(1/ω_1)PV(1/ω_2)δ(ω_3), which do not obviously vanish after the integrations over l_1^0 and l_3^0 in Eq. (20). The sentence 'As is also inferred from energy conservation' does not justify discarding the principal-value part, because energy conservation is already enforced by the spatial delta functions in Eq. (20) and does not fix the distributional definition of the time integrals. Since no argument is given that the principal-value contributions cancel, Eq. (30) is not derived from QCD; it is built into the chosen regularization of I(ω).
[Sec. 4, normalization β] The overall normalization β=2θ(0) is not determined by the QCD derivation. The author states this explicitly and fixes β=1 by demanding that Eq. (30) match the pole term of Eq. (1) in the WZW low-energy effective theory. This is a circular validation of the claim: the size of the claimed pion-pole contribution is taken from the very formula whose principal-value part is being questioned. Without an independent derivation of β from QCD, Eq. (30) cannot be used to conclude that the standard pion-pole formulas are inapplicable, since the only quantitative comparison made to those formulas is used as input.
[Sec. 2, Eq. (10) and Sec. 4, Eq. (35)] The definition of the 'neutral pion intermediate state contribution' is representation-dependent. The derivation singles out the pion from one particular insertion of a complete set of states, the first way in Eq. (9), and the author acknowledges in Sec. 4, around Eq. (35), that averaging over different insertions yields a different decomposition in which no single neutral pion term appears. The paper does not show that the neutral pion contribution defined in Eq. (13) is independent of the chosen insertion, nor that the newly proposed two-pion contribution of Eq. (35) is consistent with the pion contribution of Eq. (30). This leaves the central physical interpretation—that there is a unique neutral pion contribution with no principal-value part—unsubstantiated.

minor comments (5)

[Abstract and Sec. 1] The phrase 'no principle value term' should read 'no principal value term'; the same typo appears in the abstract and in the text near Eq. (31).
[Eq. (20)] The notation in Eq. (20) is extremely dense; the delta functions and the step-function constraint are written inline, making it hard to verify the domain of integration for the time variables. A clearer display of the integrand would improve readability.
[Sec. 4, Eq. (35)] In the last two lines of Eq. (35), the matrix element ⟨0|𝒯[𝐽_{ν(1)}(z(1))𝐽_{ν(1)}(z(2))𝐽_ν(0)]|v⟩ appears with two identical Lorentz indices ν(1); this is presumably a typo and should involve ν(3) or another index.
[References] Reference [9] lists the page number as '9' in 'Phys. Rev. D 108, no.9, 9'; the article identification number should be given (e.g., 094514). The preprint number field at the top is left as 'XXXX-XXXX'.
[Sec. 3, Eq. (24)] The symbol a is introduced and then determined in Eq. (25), but the proportionality constant is later redefined as β=2θ(0); this two-step notation is confusing and could be streamlined.

Circularity Check

2 steps flagged · score 8.0 of 10

The no-principal-value claim is inserted by hand in Eq. (24), and the normalization is fixed by WZW matching to the very Eq. (1) whose pathology is at issue.

self definitional [Sec. 3, Eqs. (21)–(25); decisive replacement (24)]
"The ambiguity arises since no definition is found for the integral ∫∞0 dte iωt. (23) The imaginary part may be set to 0 by dividing a half line into the intervals of length 2πω−1 on each of which the integral of sin(ωt) vanishes. The same consideration is applied for the real part except for ω=0. As is also inferred from energy conservation, we set I(ω)=a 3∏ j=1 δ(ωj) (24) with some constant a."

The central claim is that Eq. (30) contains only the delta-function/pole part and no principal-value contribution. That conclusion is exactly what Eq. (24) assumes: the ambiguous iterated time integral I(ω) is defined to be a product of δ-functions, which has no PV part. The standard distributional identity for ∫0∞ dt e−iωt is πδ(ω) − i PV(1/ω) (up to convention); using it in Eq. (22) generates many PV terms after the l10 and l30 integrations. The paper gives no derivation from QCD or from the already-enforced spatial delta functions that would eliminate those PV terms. Energy conservation in Eq. (20) fixes the argument of the δ-functions, not the distributional meaning of the product of half-line integrals. Thus the absence of PV is put in by definition at Eq.
fitted input called prediction [Sec. 1 (before Eq. (5)) and Sec. 4 (after Eq. (31))]
"The approach of this work is unable to deduce the overall normalization β for the neutral pion contribution (30). Here, the low energy effective theory is adopted as a guiding principle to fix it : β=1. ... The demand that Eq. (30) equals the pole term in Eq. (1) in this approximation leads β=1 (θ(0)=1/2)."

The overall strength of the claimed neutral-pion contribution is not derived from the intermediate-state calculation: it is fixed by demanding that Eq. (30) reproduce the imaginary part of Eq. (1) as computed in the WZW low-energy effective theory. But Eq. (1) is precisely the standard pion-pole formula whose validity and pathology the paper is re-examining. The normalization of the 'predicted' contribution is therefore imported from the target formula, while the only structurally new feature (no PV) is already inserted through Eq. (24). The result (30) with β=1 is thus constructed to match the pole part of Eq. (1), rather than independently predicting it.

full rationale

Up to Eq. (21), the paper's intermediate-state insertion and Fourier algebra are self-contained and do not rely on load-bearing self-citations: Refs. [11,12] are lattice benchmark papers, not used to justify the central derivation. The circularity sits in two places. First, the decisive manipulation is Eq. (24), where the ambiguous half-line integral ∫0∞ dt e^{iωt} is replaced by a pure δ-function normalization. This is a nonstandard definition, not a consequence of QCD, and it already excludes the principal-value part whose absence is the paper's main claim. With the standard identity the PV terms survive, so Eq. (30) is not established as a derivation. Second, the overall factor β is admitted to be undeducible and is fixed by matching the WZW pole term of Eq. (1), the very formula under scrutiny. Thus the central 'no principal-value contribution' result is forced by a regularization convention chosen in the setup, and the size of the surviving pole term is fitted to the standard formula. This makes the main claim circular by construction rather than an independent field-theoretic derivation.

Assumptions & free parameters 1 free parameters · 3 assumptions · 0 invented entities

The central result is fixed by two non-derivable choices: the regularization of the ambiguous integral, which eliminates the principal-value term, and the matching of β to the WZW effective theory, which sets the overall normalization. No new particles or forces are introduced.

free parameters (1)

β (or θ(0)) = 1
The derivation leaves an overall normalization β=2θ(0) undetermined (Eq. 25); it is set to 1 by matching to the leading-order WZW low-energy effective theory in Sec. 4.

assumptions (3)

domain assumption The four-current correlator may be decomposed by inserting a complete set of QCD states with the specific 'first way' of Eq. (9) for all 24 time orderings.
The result depends on this choice; a different choice (Eq. 35) is acknowledged in Sec. 4 to yield no single-pion contribution.
ad hoc to paper The integral I(ω) is defined so that ∫_0^∞ dt e^{iωt} yields a delta function only, with no principal-value part.
Standard distribution theory gives πδ(ω) + i PV(1/ω); the paper's choice directly removes the principal-value terms.
domain assumption Time-reversal and parity symmetries of QCD with vanishing θ are used to relate ⟨h|TJJ|0⟩ to the transition form factor.
Invoked in Sec. 3 around Eq. (16); standard but relies on the neutral pion being a stable single-particle state in pure QCD.

how reviews work

0 comments

Cite this review

Pith. "Pith review of Field-theoretical approach to neutral pion contribution to muon $g-2$." pith.science (2026). https://pith.science/paper/T54X2EQY

@misc{pith2026250116547,
  author       = {Pith},
  title        = {Pith review of: Field-theoretical approach to neutral pion contribution to muon $g-2$},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/T54X2EQY}},
  note         = {Machine review of arXiv:2501.16547}
}

abstract

The hadronic light-by-light scattering induces a substantial contribution to the muon $g-2$ with present accuracy of its measurement. The effect caused by the neutral pion through this scattering on the muon $g-2$ has been scrutinized intensively by model calculation and lattice QCD simulation. All of those estimates have been done based on one formula, but it has one pathological aspect. The purpose of this article is to resolve such a pathology from a field-theoretical approach, i.e., by defining the effect in question as the neutral pion intermediate state contribution to the QCD vacuum expectation value of time-ordered product of four electromagnetic currents. It leads (1) the pion pole term corresponding to the imaginary part of the pion propagator $\frac{1}{q^2 - m_\pi^2 + \mathrm{i} \epsilon}$, (2) and no principle value term corresponding to its real part, resolving the pathology.

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

16 extracted references · 1 canonical work pages

[1]

Kinoshita, B

T. Kinoshita, B. Nizic and Y. Okamoto, Hadronic Contributions to the Anomalous Magnetic Moment of the Muon , Phys. Rev. D 31, 2108 (1985). https://doi:10.1103/PhysRevD.31.2108

work page doi:10.1103/physrevd.31.2108 1985
[2]

Hayakawa, T

M. Hayakawa, T. Kinoshita and A. I. Sanda, Hadronic light by light scattering effect on muon 𝑔− 2, Phys. Rev. Lett. 75, 790-793 (1995). https://doi:10.1103/PhysRevLett.75.790 [arXiv:hep-ph/9503463 [hep-ph]]

arXiv 1995
[3]

Hayakawa, T

M. Hayakawa, T. Kinoshita and A. I. Sanda, Hadronic light by light scattering contribution to muon𝑔− 2, Phys. Rev. D54, 3137-3153 (1996). https://doi:10.1103/PhysRevD.54.3137 [arXiv:hep-ph/9601310 [hep-ph]]

arXiv 1996
[4]

Bijnens, E

J. Bijnens, E. Pallante and J. Prades, Hadronic light by light contributions to the muon 𝑔− 2 in the large N(c) limit , Phys. Rev. Lett.75, 1447-1450 (1995). [erratum: Phys. Rev. Lett.75, 3781 (1995)] https://doi:10.1103/PhysRevLett.75.1447 [arXiv:hep-ph/9505251 [hep-ph]]

arXiv 1995
[5]

Bijnens, E

J. Bijnens, E. Pallante and J. Prades, Analysis of the hadronic light by light contributions to the muon 𝑔− 2, Nucl. Phys. B 474, 379-420 (1996). https://doi:10.1016/0550-3213(96)00288-X [arXiv:hep-ph/9511388 [hep-ph]]

arXiv 1996
[6]

G ´erardin, H

A. G ´erardin, H. B. Meyer and A. Nyffeler, Lattice calculation of the pion transition form factor 𝜋0→𝛾∗𝛾∗, Phys. Rev. D 94, no.7, 074507 (2016). https://doi:10.1103/PhysRevD.94.074507 [arXiv:1607.08174 [hep-lat]]

arXiv 2016
[7]

G ´erardin, H

A. G ´erardin, H. B. Meyer and A. Nyffeler, Lattice calculation of the pion transition form factor with 𝑁 𝑓 = 2+ 1 Wilson quarks, Phys. Rev. D100, no.3, 034520 (2019). https://doi:10.1103/PhysRevD.100.034520 [arXiv:1903.09471 [hep-lat]]

arXiv 2019
[8]

G ´erardin, W

A. G ´erardin, W. E. A. Verplanke, G. Wang, Z. Fodor, J. N. Guenther, L. Lellouch, K. K. Szabo and L. Varnhorst, Lattice calculation of the 𝜋0, 𝜂 and 𝜂′ transition form factors and the hadronic light-by-light contribution to the muon 𝑔− 2, [arXiv:2305.04570 [hep-lat]]

arXiv

Show all 16 references

[9]

Alexandrou et al

C. Alexandrou et al. [Extended Twisted Mass], Pion transition form factor from twisted-mass lattice QCD and the hadronic light-by-light 𝜋0-pole contribution to the muon 𝑔− 2, Phys. Rev. D 108, no.9, 9 (2023). https://doi:10.1103/PhysRevD.108.094514 [arXiv:2308.12458 [hep-lat]]

2023 arXiv
[10]

T. Lin, M. Bruno, X. Feng, L. C. Jin, C. Lehner, C. Liu and Q. Y. Luo, Lattice QCD calculation of the𝜋0-pole contribution to the hadronic light-by-light scattering in the anomalous magnetic moment of the muon, [arXiv:2411.06349 [hep-lat]]

arXiv
[11]

T. Blum, N. Christ, M. Hayakawa, T. Izubuchi, L. Jin, C. Jung and C. Lehner, Hadronic Light-by-Light Scattering Con- tribution to the Muon Anomalous Magnetic Moment from Lattice QCD , Phys. Rev. Lett. 124, no.13, 132002 (2020). https://doi:10.1103/PhysRevLett.124.132002 [arXiv...

2020 arXiv
[12]

T. Blum, N. Christ, M. Hayakawa, T. Izubuchi, L. Jin, C. Jung, C. Lehner and C. Tu, Hadronic light-by-light contribution to the muon anomaly from lattice QCD with infinite volume QED at physical pion mass , Phys. Rev. D 111, no.1, 014501 (2025). https://doi:10.1103/PhysRevD.11...

2025 arXiv
[13]

E. H. Chao, R. J. Hudspith, A. G ´erardin, J. R. Green, H. B. Meyer and K. Ottnad, Hadronic light-by-light contribution to (𝑔− 2)𝜇 from lattice QCD: a complete calculation, Eur. Phys. J. C 81, no.7, 651 (2021). https://doi:10.1140/epjc/s10052-021-09455-4 [arXiv:2104.02632 [hep-lat]]

2021 arXiv
[14]

Colangelo, M

G. Colangelo, M. Hoferichter, M. Procura and P. Stoffer, Dispersive approach to hadronic light-by-light scattering, JHEP 09, 091 (2014). https://doi:10.1007/JHEP09(2014)091 [arXiv:1402.7081 [hep-ph]] 17

2014 arXiv
[15]

Knecht and A

M. Knecht and A. Nyffeler, Hadronic light by light corrections to the muon𝑔− 2: The Pion pole contribution, Phys. Rev. D 65, 073034 (2002). https://doi:10.1103/PhysRevD.65.073034 [arXiv:hep-ph/0111058 [hep-ph]]

2002 arXiv
[16]

Colangelo, M

G. Colangelo, M. Hoferichter, M. Procura and P. Stoffer, Hadronic light-by-light contribution to (𝑔− 2)𝜇: a dispersive approach, EPJ Web Conf. 175, 01025 (2018). https://doi:10.1051/epjconf/201817501025 [arXiv:1711.00281 [hep-ph]] 18

2018

Pith tools

Reviewed August 10, 2026 · model on record in the stance chip above.

[1] [1]

Kinoshita, B

T. Kinoshita, B. Nizic and Y. Okamoto, Hadronic Contributions to the Anomalous Magnetic Moment of the Muon , Phys. Rev. D 31, 2108 (1985). https://doi:10.1103/PhysRevD.31.2108

work page doi:10.1103/physrevd.31.2108 1985

[2] [2]

Hayakawa, T

M. Hayakawa, T. Kinoshita and A. I. Sanda, Hadronic light by light scattering effect on muon 𝑔− 2, Phys. Rev. Lett. 75, 790-793 (1995). https://doi:10.1103/PhysRevLett.75.790 [arXiv:hep-ph/9503463 [hep-ph]]

arXiv 1995

[3] [3]

Hayakawa, T

M. Hayakawa, T. Kinoshita and A. I. Sanda, Hadronic light by light scattering contribution to muon𝑔− 2, Phys. Rev. D54, 3137-3153 (1996). https://doi:10.1103/PhysRevD.54.3137 [arXiv:hep-ph/9601310 [hep-ph]]

arXiv 1996

[4] [4]

Bijnens, E

J. Bijnens, E. Pallante and J. Prades, Hadronic light by light contributions to the muon 𝑔− 2 in the large N(c) limit , Phys. Rev. Lett.75, 1447-1450 (1995). [erratum: Phys. Rev. Lett.75, 3781 (1995)] https://doi:10.1103/PhysRevLett.75.1447 [arXiv:hep-ph/9505251 [hep-ph]]

arXiv 1995

[5] [5]

Bijnens, E

J. Bijnens, E. Pallante and J. Prades, Analysis of the hadronic light by light contributions to the muon 𝑔− 2, Nucl. Phys. B 474, 379-420 (1996). https://doi:10.1016/0550-3213(96)00288-X [arXiv:hep-ph/9511388 [hep-ph]]

arXiv 1996

[6] [6]

G ´erardin, H

A. G ´erardin, H. B. Meyer and A. Nyffeler, Lattice calculation of the pion transition form factor 𝜋0→𝛾∗𝛾∗, Phys. Rev. D 94, no.7, 074507 (2016). https://doi:10.1103/PhysRevD.94.074507 [arXiv:1607.08174 [hep-lat]]

arXiv 2016

[7] [7]

G ´erardin, H

A. G ´erardin, H. B. Meyer and A. Nyffeler, Lattice calculation of the pion transition form factor with 𝑁 𝑓 = 2+ 1 Wilson quarks, Phys. Rev. D100, no.3, 034520 (2019). https://doi:10.1103/PhysRevD.100.034520 [arXiv:1903.09471 [hep-lat]]

arXiv 2019

[8] [8]

G ´erardin, W

A. G ´erardin, W. E. A. Verplanke, G. Wang, Z. Fodor, J. N. Guenther, L. Lellouch, K. K. Szabo and L. Varnhorst, Lattice calculation of the 𝜋0, 𝜂 and 𝜂′ transition form factors and the hadronic light-by-light contribution to the muon 𝑔− 2, [arXiv:2305.04570 [hep-lat]]

arXiv