{"id":"99fbc9ba-0af6-4dd8-adc6-3d4dd0238249","arxiv_id":"2504.13588","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":5.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":4,"one_line_summary":"The nonlocal chiral quark model with vector and axial-vector mesons yields a hadronic light-by-light contribution to the muon g-2 of (157±10.6)x10^-11, with vector-meson photon dressing contributing only a few units.","lead":"Physicists extended a quark model calculation of the hadronic light-by-light contribution to the muon's magnetic moment by adding vector and axial-vector mesons. They find the new vector-meson effects are tiny, leaving a total estimate of 157 plus or minus 10.6 in units of 10^-11, close to earlier model values but higher than lattice QCD results.","discovery_kind":"extension","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The quoted total in Eq. (16) is not a complete model prediction: the pion-loop (1/Nc) contribution is acknowledged as missing and negative, and the strange sector is imported from [28]; either could move the central value beyond the stated uncertainty.","rationale":"The paper's novel numerical result—that vector-meson dressing of internal photons is tiny—is supported by explicit diagram-by-diagram numbers and by the local-limit cross-checks in Appendix B, so I do not question it. The central claim that needs scrutiny is the quoted total, Eq. (16). For that total to be a meaningful model prediction, every significant contribution in the model must be included or its omission quantified. The paper explicitly omits the pion loop and adds the whole strange sector from the earlier no-vector model. The reader's listed weakest assumption (strange-sector reuse) is real but numerically minor: the strange pieces contribute only about 7×10^-11 to the total. The pion-loop omission is far more consequential: the paper itself states it is a known negative contribution and a source of the discrepancy with dispersive and lattice estimates. A negative shift of tens of units would bring 157 in line with the roughly 100–130 values in Fig. 9, which is exactly the size of effect needed. Therefore, while the vector-dressing part of the central claim survives, the 'total' should be presented as a leading-1/Nc partial estimate until the pion loop is computed or bounded. The abstract/body error-bar mismatch for the quark loop (15.4 vs 9.1) only reinforces that the quoted uncertainty is not yet a reliable statement. My verdict matches the reader's: conditional acceptance, with the condition that the missing pion loop and the error-bar discrepancy be addressed. No change to the reader's verdict is therefore needed.","tokens_in":15128,"tokens_out":9077,"duration_ms":86040,"concrete_test":"Quantify the pion-loop contribution in the same nonlocal framework, e.g., by implementing the 1/Nc meson-loop scheme of Ref. [54] with the model's nonlocal form factors, or by evaluating the two-pion two-photon box with those form factors. If the resulting negative shift is smaller than ±10.6, Eq. (16) can stand as a total; if it is of order tens of 10^-11, the central value should be revised down and the comparison in Fig. 9 redrawn.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The most load-bearing weakness is the status of Eq. (16) as a 'total'. Section V adds the entire strange sector (η, η′, f0(980), a0(980), strange quark loop) from the model without spin-1 mesons [28] without recomputation; Section VI then states that the mean-field calculation does not include the pion-loop contribution, a subleading 1/Nc term, and that this contribution is known to be negative [7,53]. The strange-sector piece is only about 7×10^-11, so even a sizeable relative change shifts the total by a few units. The pion loop is different: the paper itself identifies it as the known reason the nonlocal model overshoots dispersive and lattice results (Fig. 9). If, as suggested there, this missing negative contribution is of order tens of 10^-11, the central value of 157 would fall by more than the quoted 10.6 uncertainty. Independently, the abstract's non-strange loop value (96±15.4)×10^-11 is inconsistent with Section V's 96±9.1, so the error budget of the quote is not internally traceable. The vector-dressing claim (-2.9, 0.09, 0.001) is explicitly computed and not in question.","agreement_with_reader":"partial"},"referee_report":{"model":"deepseek-v4-flash","summary":"This paper computes the hadronic light-by-light (HLbL) contribution to the muon anomalous magnetic moment in the nonlocal chiral quark model extended with vector and axial-vector mesons. The main new effect, the dressing of internal photon lines by intermediate ρ(ω) exchanges, is found to be tiny: at md = 310 MeV the one-, two-, and three-vector-meson-exchange contributions are -2.9, 0.09, and 0.001 (in units of 10^-11), respectively. The non-strange quark loop is reported as (96±9.1)×10^-11 in Section V (the abstract quotes 96±15.4), the SU(2) total including meson exchanges is 150±9.6, and adding strange-particle contributions from Ref. [28] gives the final result a_mu^HLbL = (157±10.6)×10^-11 in Eq. (16). The result is compared with lattice, dispersive, DSE, and AdS/QCD determinations in Fig. 9.","tokens_in":15388,"tokens_out":4313,"duration_ms":35143,"significance":"If the quoted result survives closer scrutiny, the paper's main message is that extending the nonlocal quark model to spin-1 mesons leaves the HLbL value essentially unchanged and that the vector-meson dressing of internal photons is negligible, in contrast to the large reduction found in the early ENJL estimate. The numerical procedure is supported by useful cross-checks in Appendix B, including local limits and comparisons with VMD and ENJL dressings, and the authors explicitly identify the main formal limitation (the missing pion loop). However, the central value in Eq. (16) is not a complete model prediction: the strange-sector contributions are imported from the model without vector mesons, and the pion-loop contribution is omitted and acknowledged to be negative. The paper is therefore a valuable model study with conditional impact on the current g-2 discussion.","major_comments":[{"comment":"The final total in Eq. (16) is obtained by adding the strange-particle contributions (η, η′, f0(980), a0(980), and the strange quark loop) from the model without spin-1 mesons [28] under the assumption that they \"will not change much\" after introducing vector–axial-vector fields. This is an uncomputed assumption: the paper does not recompute these diagrams in the V-AV model or estimate the associated uncertainty. Since these contributions amount to roughly 7×10^-11, a moderate relative change would shift the central value of Eq. (16) by more than the quoted ±10.6 uncertainty. The total should either be recomputed in the extended model or presented with an explicit caveat and an appropriate uncertainty.","section":"Section V, after Fig. 8"},{"comment":"The paper explicitly states that the mean-field calculation does not include the pion-loop contribution, a subleading 1/Nc term, and that this contribution is known to be negative [7,53]. Because the pion loop can plausibly shift the result by tens of 10^-11, the value 157±10.6 in Eq. (16) is not a complete model prediction and the comparison with lattice and dispersive values in Fig. 9 is incomplete. The authors acknowledge this limitation in the text, but the abstract and Section V present Eq. (16) without this caveat. An estimate of the pion-loop contribution, or at least a quantitative bound, is needed before the total can be meaningfully compared with other determinations.","section":"Section VI, final paragraphs"},{"comment":"The abstract quotes the non-strange quark loop as (96±15.4)×10^-11, while Section V reports 96±9.1 for the same quantity at md = 310 MeV. The error 15.4 does not appear anywhere in the body of the paper; the text instead gives 105±9.1 for the range-averaged loop and 96±9.1 for the chosen point. This inconsistency makes the error budget of the headline numbers internally untraceable and must be corrected.","section":"Abstract and Section V"},{"comment":"The quoted uncertainties for the central values at md = 310 MeV (9.1 for the loop, 9.6 for the SU(2) total) appear to be inherited from the range-averaging procedure over md from 200 to 350 MeV rather than derived at the chosen point. At md = 310 MeV the spread between the G2/G1 = -0.08 and -0.14 curves is about 3 units for the quark loop, so the origin of the ±9.1 error bar is unclear. The paper should state explicitly how each error quoted in Eq. (16) and in the preceding values is computed.","section":"Section V, uncertainty derivation"}],"minor_comments":[{"comment":"In the sentence \"the contributions to the ratio of −0.08 from quark loops and mesons are 99 and 51,\" the phrase \"the ratio of −0.08\" should likely read \"the parameter set with G2/G1 = −0.08\"; please rephrase for clarity.","section":"Section V, Fig. 8 discussion"},{"comment":"In the sentence \"the value of the vector meson correction will be small if Λ ∼ 1 GeV,\" the word \"if\" should likely be \"for\"; please check the intended meaning.","section":"Appendix B"},{"comment":"Equation (1) is typeset in a way that is hard to read, with the coupling constants and field terms not cleanly separated; please re-typeset to improve readability.","section":"Section II, Eq. (1)"}],"recommendation":"major_revision","confidential_remarks":"The paper's central numerical result rests on two acknowledged approximations, namely the imported strange-sector contributions and the missing pion loop, neither of which is included in the quoted uncertainty. The authors should be encouraged to either compute these pieces within the extended model or clearly label Eq. (16) as a mean-field estimate with a corresponding caveat. The cross-checks in Appendix B are valuable and should be retained in the revised version."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Dear colleague,\n\nThe genuinely new thing here is the vector-meson dressing calculation. In the nonlocal chiral quark model with spin-1 mesons, the effect of dressing internal photons by one, two, and three rho/omega exchanges is -2.9, 0.09, and 0.001 (in 10^-11 units) at md=310 MeV. That is an explicit, computed statement, and it supports the paper's main qualitative claim that the effect is tiny. The paper also does honest cross-checks (local limits, ENJL/VMD comparison) and does not fit anything to HLbL. Credit where due.\n\nThe soft spot is the headline total, not the vector dressing. Three things bother me. First, the abstract quotes the non-strange quark loop as (96±15.4)x10^-11, while the body says 96±9.1; same number, different error. That inconsistency makes the error budget untraceable and needs fixing before the paper is usable. Second, the strange-sector contributions (eta, eta-prime, f0, a0, strange quark loop) are imported from the earlier model without spin-1 mesons [28] and added 'naively'. The paper explicitly assumes they do not change much, but it does not recompute them or assign an uncertainty to that assumption. The strange sector is only about 7x10^-11, so this is not fatal, but it is a real gap in a claimed total. Third, and most important, the pion-loop contribution is missing. The paper says so in Section VI: the mean-field calculation does not include it, and it is known to be negative. If it is of order tens of 10^-11, as the comparison with dispersive and lattice results in Fig. 9 suggests, then Eq. (16) is not a complete model prediction, and the central value could fall by more than the quoted ±10.6. The authors are upfront about this, which I respect, but it means the abstract overstates the status of the result.\n\nI do not think the core calculation is wrong. The vector dressing numbers are explicit and the model framework is established. The citation pattern is fine: they build on their own [28,29] and cite the relevant lattice, dispersive, ENJL and DSE literature. There is no circularity—no parameter is fit to HLbL.\n\nWho gets value from this: people who want another effective-model cross-check for HLbL, especially the size of vector-meson corrections. It will not resolve the g-2 tension, and it should not be quoted as a definitive number until the error inconsistency is fixed and the pion-loop issue is addressed or at least quantified.\n\nMy recommendation: send it to peer review. A serious referee can handle the presentation fixes and push the authors to either recompute the strange sector or clearly label Eq. (16) as a partial estimate. It deserves referee time, but the published version should not carry the current abstract.","headline":"The vector-meson dressing calculation is solid and the effect is tiny, but the headline total is not yet a complete prediction: the error bar is internally inconsistent, the strange sector is imported from an older model, and the negative pion loop is missing.","tokens_in":15922,"tokens_out":3097,"would_cite":true,"duration_ms":28488,"reading_group":"yes","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":[],"pacs":[],"model":"deepseek-v4-flash","headline":"Including vector and axial-vector mesons leaves the muon g-2 hadronic light-by-light term at $(157\\pm10.6)\\times10^{-11}$.","keywords":["muon g-2","hadronic light-by-light","nonlocal chiral quark model","vector meson dressing","axial-vector mesons","anomalous magnetic moment","quark loop","rho-omega transition"],"falsifier":"Recompute the strange quark loop and the $\\eta$, $\\eta'$, $f_0(980)$, and $a_0(980)$ exchange contributions with the vector- and axial-vector-dressed vertices used elsewhere in this paper; if that block shifts by more than about five units, the quoted $157\\pm10.6$ is no longer the model's answer. Alternatively, a lattice or dispersion measurement of the hadronic light-by-light contribution below about $135\\times10^{-11}$ with an uncertainty below $5$ would exclude the central value at more than two standard deviations.","tokens_in":14886,"feed_emoji":"⚛️","tokens_out":11746,"duration_ms":100630,"temperature":0.7,"pith_summary":"This paper extends a nonlocal chiral quark model to include vector and axial-vector mesons and computes the hadronic light-by-light contribution to the muon's anomalous magnetic moment. It reports a total of $a_\\mu^{\\mathrm{HLbL}}=(157\\pm10.6)\\times10^{-11}$, close to the $(168\\pm12.5)\\times10^{-11}$ obtained earlier without spin-1 mesons. The new effect, the dressing of internal photon lines by vector-meson transitions, is tiny: at the preferred parameter point it contributes about $-2.9\\times10^{-11}$, with two- and three-exchange terms below $0.1\\times10^{-11}$. If this result holds, the model's hadronic light-by-light prediction is stable under adding spin-1 mesons, and the remaining Standard Model discrepancy with the measured muon anomaly must be accounted for by other hadronic pieces, chiefly hadronic vacuum polarization.","feed_headline":"Adding vector mesons barely moves muon g-2 hadronic term","feed_subtitle":"Vector-meson photon dressing shifts the nonstrange loop by under three units, keeping the model's answer near 157.","key_machinery":"The calculation rests on gauge-invariant nonlocal vertices generated by a path-ordered exponential phase factor attached to each quark field, which produces quark-antiquark-photon and meson-quark-antiquark-photon vertices with any number of photons. In the vector sector, every photon-quark vertex is dressed by a transverse $\\rho(\\omega)\\to\\gamma$ transition encoded in the function $C_{\\gamma V}(q^2)$ together with the transverse projector; since $C_{\\gamma V}(0)=0$, the vector mesons appear only on internal lines, and the light-by-light amplitude receives at most three vector-meson exchanges. The four-rank polarization tensor is built from the quark box plus the nonlocal multiphoton vertices, and the muon anomaly is extracted with the standard projection formula reduced to a three-dimensional integral. A local limit with constituent quark mass $m_Q=300$ MeV serves as a cross-check, recovering the expected combinatorial coefficients $-3$, $3$, and $-1$ for one-, two-, and three-exchange diagrams.","core_discovery":"The paper establishes that, in the nonlocal chiral quark model with scalar-pseudoscalar and vector-axial-vector channels, the hadronic light-by-light contribution to the muon anomalous magnetic moment is $a_\\mu^{\\mathrm{HLbL}}=(157\\pm10.6)\\times10^{-11}$. The non-strange quark loop alone is about $96\\times10^{-11}$; adding the $\\pi$, $\\sigma$, $a_1$, and $f_1$ exchange contributions gives an SU(2) value of $150\\pm9.6$ at the preferred dynamical mass $m_d=310$ MeV, where both the $\\rho$ and $a_1$ masses can be reproduced. The strange-particle pieces ($\\eta$, $\\eta'$, $f_0(980)$, $a_0(980)$, and the strange quark loop) are added from the earlier model without spin-1 mesons. The genuinely new ingredient, the dressing of internal photons by $\\rho(\\omega)\\to\\gamma$ transitions, contributes only $-2.9$, $0.09$, and $0.001$ (in units of $10^{-11}$) for one, two, and three vector-meson exchanges at that point.","pith_inferences":["Editorial inference: recomputing the strange-sector contributions inside the extended model is the immediate test; a shift of even a few units in that block would move the quoted total outside its error bar.","Editorial inference: the rapid convergence of the one-, two-, and three-exchange series suggests that truncating at one vector-meson exchange is safe for other observables in this model, such as hadronic vacuum polarization.","Editorial inference: the larger central value compared with lattice and dispersive results points to the missing pion-loop contribution, a $1/N_c$ correction expected to be negative, as the most likely single mechanism to bring the prediction down; calculating that loop in the nonlocal framework would be decisive.","Editorial inference: the local-limit cross-check, in which all vector-meson exchanges combine to minus the quark loop, could be turned into a general consistency condition for any nonlocal model with vector mesons."],"forward_implications":["If the quoted total is right, the earlier scalar-pseudoscalar estimate of $168\\times10^{-11}$ was already within about one standard deviation of the extended-model value, so the nonlocal framework is stable under this extension.","Vector-meson dressing of internal photons is a sub-3-unit effect, so future calculations in the same model can safely keep only the quark loop and one-vector-meson corrections.","Including the axial-vector mesons $a_1$ and $f_1$ does not destabilize the sum; near $m_d=310$ MeV the total is nearly independent of the ratio $G_2/G_1$.","The central value $157\\times10^{-11}$ is larger than recent lattice and dispersive determinations; if the newer lattice-based hadronic vacuum polarization is used, the residual gap to experiment narrows no matter which light-by-light value is adopted."],"supporting_citations":[{"why":"Supplies the previous no-vector-meson hadronic light-by-light value, its parameter strategy, and the strange-particle contributions that are added to the new total.","marker":"[28]"},{"why":"Provides the refitted model parameters, the $G_1/G_2$ range, and the $m_d=310$ MeV point where the $\\rho$ and $a_1$ masses are reproduced.","marker":"[29]"},{"why":"Defines the nonlocal currents and Gaussian form factor motivated by the instanton liquid model used throughout the calculation.","marker":"[31]"},{"why":"Gives a comparison estimate in which vector-meson dressing suppresses the light-by-light result by about a factor three, the contrast for the tiny effect found here.","marker":"[19]"},{"why":"Gives a comparison estimate in which the dressing changes the result by only a few percent, another benchmark for the present small correction.","marker":"[30]"},{"why":"Supplies the contour-integral rules for the path-ordered phase factor that generate the nonlocal multiphoton vertices.","marker":"[36]"},{"why":"Establishes the transversality of the vector-meson-photon transition and $C_{\\gamma V}(0)=0$, used to restrict exchanges to internal lines.","marker":"[38]"},{"why":"Shows the vector-meson transition does not renormalize the photon mass or quark charge, justifying the dressing treatment.","marker":"[39]"},{"why":"Gives the projection formula used to extract the muon anomaly from the four-rank polarization tensor.","marker":"[40]"}],"fun_headline_variants":["Vector mesons shift muon g-2 hadronic term by less than 3","Nonlocal quark model pins muon g-2 light-by-light at 157","Tiny vector-meson dressing leaves muon g-2 loop near 96","Dressed photons barely nudge muon g-2 hadronic value","Muon g-2: vector mesons add just -2.9 to light-by-light"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The final total assumes that the strange-quark and strange-meson contributions ($\\eta$, $\\eta'$, $f_0(980)$, $a_0(980)$, strange loop) computed in the model without vector and axial-vector mesons stay essentially unchanged when spin-1 mesons are introduced, and can simply be added on.","fun_headline_variants_meta":{"raw":{"variants":["Vector mesons shift muon g-2 hadronic term by less than 3","Nonlocal quark model pins muon g-2 light-by-light at 157","Tiny vector-meson dressing leaves muon g-2 loop near 96","Dressed photons barely nudge muon g-2 hadronic value","Muon g-2: vector mesons add just -2.9 to light-by-light"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000675,"raw_usage":{"total_tokens":3058,"prompt_tokens":915,"completion_tokens":2143,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":531,"completion_tokens_details":{"reasoning_tokens":2035}},"tokens_in":531,"tokens_out":2143,"duration_ms":14178,"temperature":1.0,"reasoning_tokens":2035,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-16T12:04:40.021870+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Recompute the strange quark loop and the $\\eta$, $\\eta'$, $f_0(980)$, and $a_0(980)$ exchange contributions with the vector- and axial-vector-dressed vertices used elsewhere in this paper; if that block shifts by more than about five units, the quoted $157\\pm10.6$ is no longer the model's answer. Alternatively, a lattice or dispersion measurement of the hadronic light-by-light contribution below about $135\\times10^{-11}$ with an uncertainty below $5$ would exclude the central value at more than two standard deviations.","supporting_citations":[{"cited_title":"Bijnens, E","cited_arxiv_id":null,"evidence_quote":"Gives a comparison estimate in which vector-meson dressing suppresses the light-by-light result by about a factor three, the contrast for the tiny effect found here."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Supplies the contour-integral rules for the path-ordered phase factor that generate the nonlocal multiphoton vertices."},{"cited_title":"Terning, Phys","cited_arxiv_id":null,"evidence_quote":"Establishes the transversality of the vector-meson-photon transition and $C_{\\gamma V}(0)=0$, used to restrict exchanges to internal lines."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Gives the projection formula used to extract the muon anomaly from the four-rank polarization tensor."}],"review_version":1}