{"id":"f03c8510-435a-4104-92e1-6aba568ec8f0","arxiv_id":"2411.10226","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":4.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":3,"one_line_summary":"Updated isospin-breaking corrections bring tau decay data into agreement with the new CMD-3 e+e- result, supporting a tau-based hadronic vacuum polarization estimate that reduces the muon g-2 tension to 2.7 sigma.","lead":"This paper checks whether measurements of two-pion production in electron-positron collisions and in tau decays can be reconciled after correcting for isospin-breaking effects. The author finds the corrections are under control and that tau data can then be used to cross-check the muon's anomalous magnetic moment, softening the current disagreement from 5.1 sigma to 2.7 sigma.","discovery_kind":"extension","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Systematic error budget only spans GS versus dispersive; GP/Seed totals differ from the quoted Delta a by ~2.4-3.0 x 10^-10, so model-selection dependence is the load-bearing weak point.","rationale":"The reader's weakest assumption identifies the same load-bearing point: the model uncertainty in |F_V/f_+|^2 is represented only by the GS-versus-dispersive difference. My reading of Tables 1 and 2 strengthens this concern with a quantitative observation: the GP and Seed parametrizations, which are also presented in the paper, give total Delta a values that differ from the quoted reference by roughly 2.4-3.0 x 10^-10, i.e., by more than the quoted total uncertainty in one direction. The paper's defense is that fits and analyticity favor GS and dispersive, but since that selection is partly based on fit quality after excluding KLOE, the model-selection step is itself data-dependent. The proposed Bayesian model-average test would show whether the excluded parametrizations are robustly disfavored or whether they should widen the quoted uncertainty. I do not recommend changing the reader's CONDITIONAL verdict; the concern reinforces the condition rather than overturning it.","tokens_in":10709,"tokens_out":5378,"duration_ms":54532,"concrete_test":"Recompute Delta a and Delta B with a Bayesian model average over the GS, KS, GP, Seed, and all dispersive variants, using the same fit likelihoods as Ref. [1], both including and excluding KLOE; report the posterior mean and 68% interval. If the model-averaged mean moves more than ~1 x 10^-10 from -15.15, or if the 68% interval includes -12.7, the GS-versus-dispersive linear systematic does not cover the model uncertainty and the 'under control' conclusion needs a larger uncertainty.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The paper's central assertion that the IB uncertainty is 'under control' rests on taking the dispersive p4-1 result as reference and adding linearly only its difference from GS (Section 4; Tables 1-2). But Tables 1 and 2 also list GP and Seed parametrizations whose total corrections disagree with the quoted reference by far more than that spread: for Delta a_had,LO_mu[pi pi, tau], GP gives -12.13 and Seed -12.74 versus the quoted -15.15, with differences of +3.02 and +2.41 x 10^-10, both exceeding the final +2.37/-2.90 error in one direction. For Delta B_CVC^pi pi, GP and Seed give +0.48 versus +0.63, a 0.15 x 10^-2 shift comparable to the quoted +/-0.09/-0.08 uncertainty. The selection of GS and dispersive is justified by fits and analyticity tests that improve when KLOE is excluded; this is a reasonable criterion, but it means the quoted model uncertainty is conditional on a data-dependent model choice. If the true model uncertainty is at the GP/Seed scale, both central values and the 2.7 sigma conclusion shift by more than the quoted uncertainty.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"This proceedings paper revisits the isospin-breaking (IB) corrections that relate the e+e- -> pi+ pi- cross section and the tau- -> pi- pi0 nu_tau decay spectrum, focusing on the ratio |F_V/f_+|^2 of the neutral and charged pion form factors. The author evaluates five form-factor parametrizations (GS, KS, GP, Seed, dispersive), selects GS and the dispersive result as reference, and adds the GS-versus-dispersive difference linearly as a model systematic. The main outputs are the IB corrections to the CVC prediction for B(tau -> pi pi nu_tau), Delta B_CVC^{pi pi} = (+0.63^{+0.09}_{-0.08}) x 10^-2, and to the tau-data-based hadronic vacuum polarization contribution, Delta a_mu^{HVP,LO}[pi pi, tau] = (-15.15^{+2.37}_{-2.90}) x 10^-10, implying a_mu^exp - a_mu^SM = (14.8^{+5.1}_{-5.4}) x 10^-10, a 2.7 sigma difference. The paper concludes that the IB uncertainty is under control and that tau data can be used for precision CVC tests and for the SM prediction of a_mu.","tokens_in":10920,"tokens_out":6038,"duration_ms":58914,"significance":"If the central claim holds, the paper provides a timely tau-data-based cross-check of the hadronic vacuum polarization contribution to the muon anomalous magnetic moment, particularly relevant after the CMD-3 measurement. The author should be credited for computing the IB corrections separately for five different form-factor parameterizations, for splitting the correction into individual physical sources, and for comparing explicitly with the recent literature (Refs. [19,22]). The result is presented with asymmetric uncertainties and the preferred reference is defined transparently. However, the significance of the paper is conditional on the model-selection criterion and on the validation deferred to the companion paper Ref. [1]; these are the load-bearing points for the quoted uncertainties and for the 2.7 sigma conclusion.","major_comments":[{"comment":"The quoted central result for Delta a_mu^{HVP,LO}[pi pi, tau] uses the Dispersive p4-1 value as reference and adds linearly only its difference from the GS value. However, Table 2 also lists GP and Seed totals of -12.13 and -12.74, which differ from the reference -15.15 by +3.02 and +2.41 (in units of 10^-10), respectively; both exceed the quoted +2.37/-2.90 uncertainty band. For Delta B_CVC^{pi pi} (Table 1), GP and Seed give +0.48 x 10^-2 versus the reference +0.63 x 10^-2, a shift of 0.15 x 10^-2 compared with the quoted +/-0.09/0.08 uncertainty. The GS-versus-dispersive spread therefore does not, by itself, cover the model variation shown in the paper unless the fit/analyticity selection criterion reported only in Ref. [1] is strong enough to exclude GP and Seed. Please either report that criterion quantitatively in this manuscript or enlarge the model uncertainty to include the GP/Seed spread.","section":"Section 4, Table 2"},{"comment":"The manuscript does not state whether the form-factor parametrizations (GS, KS, GP, Seed, dispersive) are fitted to e+e- data, tau data, or both. Since the same R_IB(s) correction is later applied to tau spectra to obtain Delta a_mu^{HVP,LO}[pi pi, tau] and to predict B(tau -> pi pi nu_tau) in Fig. 2, any use of the tau spectral data in the fits introduces a correlation or partial circularity that is not quantified. Please specify the data sets entering each fit and, if tau data are included, provide a version of the correction obtained with fits to e+e- data alone or estimate the induced shift in the central values.","section":"Section 3"},{"comment":"The preference for GS and Dispersive is justified by the statements that fits to data and analyticity tests work best for these parameterizations and work better when KLOE is excluded, with all details deferred to Ref. [1]. Because the model-selection step is load-bearing for the uncertainty budget, the manuscript should at least present summary fit-quality and analyticity-test information (for example chi^2/dof or p-values for each parameterization, with and without KLOE). Without this information, the reader cannot verify that the selection criterion is not data-dependent in a way that biases the quoted central values and the resulting 2.7 sigma conclusion.","section":"Section 4"}],"minor_comments":[{"comment":"In the concluding paragraph, 'a 2.7 sigma difference, with agrees nicely with Refs. [19,22]' should read 'which agrees nicely with Refs. [19,22]'.","section":"Section 5"},{"comment":"The citation for the final-state radiation factor FSR is given as Ref. [35], which is Drees and Hikasa, 'Scalar top production in e+e- annihilation'; this appears unrelated to pion final-state radiation. Please check and replace the citation.","section":"Section 2, Ref. [35]"},{"comment":"The uncertainty notation in the tables (e.g. -0.08(0)(120), +0.63(86)(03), -11.96(0.15)) is not defined in the text. Please define explicitly which parentheses denote statistical, systematic, or additional model uncertainties, or use labeled subscripts so the reader can interpret the entries.","section":"Tables 1 and 2"},{"comment":"The entry labeled 'CMD3 23 IB from Davier et al. 09' appears to be an external comparison point rather than a result of this paper; please clarify whether it is shown only for comparison and which IB corrections it uses.","section":"Figure 2"}],"recommendation":"major_revision","confidential_remarks":"The manuscript is a proceedings contribution that relies heavily on the companion Ref. [1] for the fits, analyticity tests, and detailed uncertainty budgets. If Ref. [1] indeed provides the missing validation, the major comments can be addressed by summarizing the relevant results and by quantifying the GP/Seed discrepancy. If Ref. [1] does not contain that validation, the central claim that the IB uncertainty is 'under control' is not supported by the present manuscript alone. The GP/Seed model dependence is the main risk to the 2.7 sigma conclusion. I also recommend checking the FSR citation before publication."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"The core of this paper is a set of updated numbers: Delta B_CVC = +0.63 x 10^-2 and Delta a_mu^had,LO[pi pi, tau] = -15.15 x 10^-10, which bring the e+e- and tau determinations into better agreement and reduce the g-2 tension to about 2.7 sigma. These are concrete, and they agree with earlier work such as Davier et al. The paper does a good job of showing the breakdown of isospin-breaking corrections across five form factor parametrizations, and it is transparent about taking the dispersive result as reference and adding its difference with Gounaris-Sakurai linearly as a systematic. The tables are clear and the comparison with published values is useful.\n\nThe soft spot is exactly what the stress-test flags. Table 2 lists GP and Seed values for Delta a at -12.13 and -12.74, which sit 2.4 to 3.0 x 10^-10 above the quoted -15.15. That difference exceeds the quoted +2.37/-2.90 uncertainty. The paper justifies preferring GS and dispersive because the fits and analyticity tests work best for those, and better when KLOE is excluded. That is a defensible criterion, but it makes the model uncertainty conditional on a data-dependent choice. If the true model uncertainty is closer to the GP/Seed spread, both central values shift by more than the quoted error and the 2.7 sigma conclusion weakens. I think the reader's conditional verdict is right, and the stress-test concern lands.\n\nThere is also a mild circularity: the form factor ratio |F_V/f_+|^2 is computed from parametrizations fitted to e+e- and tau data, so using the corrected tau spectrum to predict a_mu is partly circular. This is standard practice and the effect may be small, but it deserves one sentence in the paper.\n\nMy main caveat is structural: every technical detail, including the fits and analyticity tests, is deferred to the companion paper [1]. For a proceedings contribution that is normal, but it means this manuscript does not stand alone. The reader who wants to rely on these numbers should go to [1].\n\nMy recommendation: this deserves peer review rather than desk rejection, but the referee should be asked to evaluate it together with the companion paper, and the author should either include the model-selection rationale or explicitly present this as a summary of [1]. For the g-2 literature, the companion is the primary reference.","headline":"Useful proceedings summary of tau-based IB corrections with CMD-3, but the 'under control' claim rests on a model-selection choice that may be too narrow.","tokens_in":11556,"tokens_out":4829,"would_cite":false,"duration_ms":43696,"reading_group":"maybe","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":[],"pacs":[],"model":"deepseek-v4-flash","headline":"The isospin-breaking corrections linking $e^+e^-$ and tau di-pion data are under control, so tau decays can serve as a competitive input for the muon g-2, with a 2.7 sigma gap between experiment and the Standard Model.","keywords":["muon g-2","hadronic vacuum polarization","pion form factor","isospin breaking","tau decay","conserved vector current","e+e- annihilation","CVC test"],"falsifier":"A sub-percent lattice QCD determination of $|F_V(s)/f_+(s)|^2$ over the $\\rho$-resonance region that deviates from the dispersive input by more than the quoted GS-versus-dispersive spread would falsify the claim that the isospin-breaking corrections are under control.","tokens_in":10395,"feed_emoji":"🧲","tokens_out":12597,"duration_ms":101578,"temperature":0.7,"pith_summary":"This paper argues that the isospin-breaking corrections that connect $e^+e^-$ annihilation into two pions to tau decays into a pion pair are under control, so that tau data can be used as a reliable input for the muon anomalous magnetic moment. The authors compare several pion form factor parametrizations and find that, although individual correction terms differ, the total correction is stable between their preferred models. With these corrections applied, tau decays predict a hadronic-vacuum-polarization contribution that leaves a $2.7\\sigma$ gap between the experimental and Standard Model values of $a_\\mu$. If the claim holds, tau data become a valuable cross-check on the $e^+e^-$ datasets that currently disagree among themselves, and precision CVC tests become feasible.","feed_headline":"Tau data can predict the muon g-2 hadronic piece","feed_subtitle":"With isospin-breaking corrections under control, tau decays give a competitive HVP input and a 2.7 sigma exp-SM gap.","key_machinery":"The central object is the isospin-breaking factor $R_{IB}(s)=[FSR(s)/G_{EM}(s)]\\,[\\beta^3_{\\pi^+\\pi^-}(s)/\\beta^3_{\\pi^+\\pi^0}(s)]\\,|F_V(s)/f_+(s)|^2$, used together with the short-distance electroweak correction $S_{EW}$ to convert tau spectra into $e^+e^-$ cross-sections. The paper concentrates on the last factor, the ratio of the neutral (electromagnetic) and charged (weak) pion form factors, where $\\rho$–$\\omega$ mixing and the neutral/charged $\\rho$ mass and width differences enter. It evaluates this ratio with five parametrizations, takes Gounaris–Sakurai and a dispersive representation (with a conformal polynomial) as reference, and adds their difference linearly as a systematic uncertainty.","core_discovery":"The authors revisit the isospin-breaking (IB) corrections relating $\\sigma(e^+e^- \\to \\pi^+\\pi^-)$ to the $\\tau^- \\to \\pi^-\\pi^0\\nu_\\tau$ spectrum, focusing on the ratio of the neutral electromagnetic form factor $F_V(s)$ to the charged weak form factor $f_+(s)$. They compare several parametrizations (Gounaris–Sakurai, Kühn–Santamaría, Guerrero–Pich, Seed, and a dispersive one) and, based on fits to data and analyticity tests, adopt the dispersive result as reference with the GS difference added linearly as a systematic. Their main results are $\\Delta B^{\\pi\\pi}_{CVC} = (+0.63^{+0.09}_{-0.08})\\times 10^{-2}$ and $\\Delta a^{had,LO}_\\mu[\\pi\\pi,\\tau] = (-15.15^{+2.37}_{-2.90})\\times 10^{-10}$, which translate into $\\Delta a_\\mu \\equiv a^{exp}_\\mu - a^{SM}_\\mu = (14.8^{+5.1}_{-5.4})\\times 10^{-10}$, a $2.7\\sigma$ difference. They conclude that the IB corrections are reliable, supporting the use of tau data in updated SM predictions of $a_\\mu$ and in precision CVC tests.","pith_inferences":["If the corrections are truly under control, the unresolved KLOE-versus-CMD-3 discrepancy in $e^+e^-$ data becomes the dominant systematic in the data-driven $a_\\mu$ determination, shifting experimental priorities toward resolving that measurement conflict.","The GS-versus-dispersive spread may underestimate model uncertainty if both parametrizations share a common bias (for example in $G_{EM}$ or the $\\rho$–$\\omega$ interference); a third independent determination, such as a lattice calculation of the form factor ratio, would test this.","A differential measurement of tau and $e^+e^-$ spectra could extract $|F_V(s)/f_+(s)|^2$ empirically at each $s$, validating the correction factor locally rather than only through integrated branching fractions.","Should the future $e^+e^-$ average be dominated by CMD-3-like data, the tau and $e^+e^-$ based values of $a_\\mu^{HVP}$ may agree within $1\\sigma$, which would recast the muon $g-2$ discrepancy as primarily an experimental data tension rather than a sign of new physics."],"forward_implications":["Tau-based determinations of $a_\\mu^{HVP}$ can now be included in the Standard Model average, providing an independent cross-check on $e^+e^-$ data that are internally inconsistent.","The CVC relation between tau and $e^+e^-$ di-pion spectra can be tested at the $\\sim 0.1\\%$ level once more precise tau spectral functions become available.","The residual $\\Delta a_\\mu = (14.8^{+5.1}_{-5.4})\\times 10^{-10}$, a $2.7\\sigma$ gap, is consistent with tau-based HVP predictions and suggests the KLOE-versus-CMD-3 disagreement in $e^+e^-$ data is the main driver of the earlier larger discrepancy.","A future high-statistics tau spectral function measurement would directly map onto the same pion form factor and can arbitrate between the competing $e^+e^-$ datasets.","The same machinery can be extended to other exclusive hadronic channels to further test CVC and reduce the uncertainty on the hadronic vacuum polarization."],"supporting_citations":[{"why":"Companion paper providing the detailed fits, analyticity tests, and uncertainty breakdown on which these proceedings' results rest.","marker":"[1]"},{"why":"Advocated the tau-based evaluation and computed the G_EM long-distance QED correction at O(p^4) in Resonance Chiral Theory, a key input to R_IB.","marker":"[12]"},{"why":"Original framework for converting tau spectral functions to e+e- observables via CVC, establishing the procedure this paper updates.","marker":"[14]"},{"why":"Revisited tau versus e+e- spectral function comparison; this paper compares against its results in Fig. 1.","marker":"[19]"},{"why":"Current landscape of e+e- data tensions; the comparison in Fig. 2 uses its approach and agrees with it.","marker":"[22]"},{"why":"Provides the width difference between neutral and charged rho mesons used in the IB corrections.","marker":"[47]"},{"why":"Dispersive representation of the pion vector form factor with conformal polynomial, used to construct the dispersive reference result.","marker":"[53]"},{"why":"CMD-3 measurement of the e+e- -> pi+pi- cross section, the new dataset used in Fig. 2 for comparison with tau predictions.","marker":"[10]"}],"fun_headline_variants":["Tau data predicts muon g-2 to 2.7 sigma gap","Isospin corrections make tau decays reliable for g-2","Tau spectrum confirms hadronic muon g-2 prediction","Improved tau input narrows muon g-2 discrepancy","Tau decays pass CVC test, support muon g-2 SM"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The conclusion rests on the assumption that the difference between the Gounaris–Sakurai and dispersive parametrizations of the pion form factor ratio spans the true model uncertainty in $|F_V(s)/f_+(s)|^2$, so that no bias larger than this spread hides in the correction.","fun_headline_variants_meta":{"raw":{"variants":["Tau data predicts muon g-2 to 2.7 sigma gap","Isospin corrections make tau decays reliable for g-2","Tau spectrum confirms hadronic muon g-2 prediction","Improved tau input narrows muon g-2 discrepancy","Tau decays pass CVC test, support muon g-2 SM"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000263,"raw_usage":{"total_tokens":1592,"prompt_tokens":930,"completion_tokens":662,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":546,"completion_tokens_details":{"reasoning_tokens":570}},"tokens_in":546,"tokens_out":662,"duration_ms":6633,"temperature":1.0,"reasoning_tokens":570,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-12T19:49:35.105600+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"A sub-percent lattice QCD determination of $|F_V(s)/f_+(s)|^2$ over the $\\rho$-resonance region that deviates from the dispersive input by more than the quoted GS-versus-dispersive spread would falsify the claim that the isospin-breaking corrections are under control.","supporting_citations":[],"review_version":1}