{"id":"498e4ed0-5754-49e5-880b-96dfe58e6046","arxiv_id":"2411.08852","paper_version":1,"verdict":"ACCEPT","confidence":"HIGH","novelty_score":5.0,"correctness_risk":"low","formal_verification":"none","parameter_count":2,"one_line_summary":"In isospin-symmetric QCD, lattice QCD gives the quark-connected strange and charm HVP contributions to the muon g-2 as 53.57(63) x 10^-10 and 14.56(13) x 10^-10.","lead":"Scientists used a supercomputer technique called lattice QCD to calculate how much strange and charm quarks contribute to the muon's magnetic moment. The new values are precise to about one percent and agree with independent lattice calculations, helping tighten the Standard Model prediction that is currently being tested by experiment.","discovery_kind":"extension","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Continuum-extrapolation ansatz completeness is the dominant risk; a leave-one-out check on the finest lattice spacing would test the quoted error.","rationale":"The calculation is careful and the external consistency with other lattice results is strong. The dominant uncertainty in the headline numbers is the continuum extrapolation, and the present AIC model set—although reasonable—does not by itself certify that the true discretization errors are contained within it. The proposed leave-one-out test exploits the newly added E112 ensemble as an independent validation point: it directly probes the model family's predictive power at the finest spacing. If the test fails, the quoted cont errors (0.48 for strange, 0.09 for charm) are not credible; if it passes, the central claim is significantly strengthened. This is a modest, concrete request that does not impugn the analysis, and it is consistent with the reader's own identification of the continuum-limit ansatz as the weakest assumption. I therefore recommend conditional acceptance pending this check.","tokens_in":35693,"tokens_out":13949,"duration_ms":130813,"concrete_test":"Remove the finest ensemble E112 (a ≈ 0.049 fm) from the data set, perform the same AIC continuum extrapolation on the remaining three spacings using all models in Eq. (15), and compute the model-averaged prediction for a_mu^HVP(s) and a_mu^HVP(c) at a = 0.04892 fm, propagating the extrapolation error and adding the statistical error of E112 in quadrature. Compare the prediction to the actual E112 values. If the discrepancy exceeds the combined error by more than 2σ, the AIC model family is incomplete and the quoted continuum errors are underestimated; if consistent, the model family passes the sharpest available test with the new finest point.","verdict_should_be":"CONDITIONAL","load_bearing_attack":"The central values in Eqs. (18)-(19) and the quoted total errors (0.63 and 0.13 in units of 10^-10) are dominated by the continuum-limit uncertainty: for the strange contribution the 'cont' error (0.48) exceeds the statistical error (0.41). The continuum limit is obtained from four lattice spacings (a ≈ 0.049-0.080 fm) using the AIC average of the model family in Eq. (15): P0 + P1 a^2 + P2 a^4, with variants where a^4 is replaced by a^2/[log(a^2/λ0^2)]^n, plus data cuts excluding the coarsest spacing(s). The AIC systematic error (Eq. (16), σ_x,syst) is the weighted spread of the central values of these models. The load-bearing assumption is that this family spans the true discretization-error space. With only four lattice spacings spanning a factor ≈2.7 in a^2, the data cannot discriminate a^2, a^4, a^2 log a^2, or residual O(a) effects (e.g., from imperfect maximal twist) which are not represented in the ansatz. The paper's checks—two regularizations (TM/OS), tmin-dependence in Sec. III A, and exclusion of coarsest points—are useful but all fit within the same parametric family, so they do not test the family's completeness. Other acknowledged limitations (charm sea-quark derivative estimated via scaling in Eq. (D2), non-blinded analysis) are smaller and explicitly quantified. Because the continuum error is the dominant component of the quoted total error, an unmodeled discretization effect could shift the central values by more than the quoted error, undermining the central claim.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"This paper presents a lattice QCD calculation of the quark-connected strange and charm hadronic-vacuum-polarization contributions to the muon anomalous magnetic moment in isospin-symmetric QCD. The analysis uses ETMC Nf=2+1+1 twisted-mass Wilson-clover ensembles at four lattice spacings (a≈0.049–0.080 fm) with volumes up to 7.6 fm, corrects for small sea- and valence-quark mass mistunings and critical-mass mistuning using reweighting and hadronic inputs from the Edinburgh/FLAG consensus, and extrapolates to the continuum using an AIC model average over a^2/a^4 and logarithmic variants. The final results are a_mu^HVP(s)=53.57(63)×10^-10 and a_mu^HVP(c)=14.56(13)×10^-10, together with SD, W, and LD window contributions, and the paper compares these with other lattice determinations.","tokens_in":36053,"tokens_out":8672,"duration_ms":223274,"significance":"If the quoted results are correct, they are among the most precise lattice determinations of the strange and charm HVP contributions and provide useful independent cross-checks for the Standard Model prediction of the muon anomalous magnetic moment. The paper is unusually transparent in its error decomposition: Eqs. (18)-(19) separate statistical, continuum, and FSE errors, the continuum extrapolation uses two current regularizations (TM/OS) and a defined AIC averaging procedure, and the Z_V and Z_A renormalization constants are determined from Ward identities with high precision. The inclusion of the new fine E112 ensemble and explicit reweighting corrections for mistunings are concrete improvements over the authors' previous work. The main residual risks are the completeness of the continuum-extrapolation model family and the scaling assumption used for the charm sea-quark mass derivative; both are acknowledged in the paper but deserve quantitative robustness checks.","major_comments":[{"comment":"The dominant error in the final results is the continuum-extrapolation uncertainty (0.48 vs 0.41 in units of 10^-10 for strange; 0.09 vs 0.10 for charm), yet this error is defined as the AIC spread within a single parametric family P0+P1^reg a^2+P2^reg a^4 and its logarithmic variants. With only four lattice spacings spanning a factor of about 2.7 in a^2, the fitted data have limited power to discriminate between these smooth forms and, for example, a residual O(a) contribution or a spacing-dependent artifact that affects mainly the finest ensemble E112. Since the model family is assumed complete, the quoted σ_cont does not include this risk. I request a leave-one-out analysis that omits the finest spacing E112 and, where possible, a fit including an explicit O(a) term, together with a discussion of how the central values in Eqs. (18)-(19) shift under these variations.","section":"Section III, Eq. (15) and Fig. 3"},{"comment":"The charm sea-quark mass derivative is estimated from the strange one by the scaling ∂_c^sea O ≃ (m_s/m_c) ∂_s^sea O because the direct derivative is too noisy. The text states that including the charm sea mistuning correction increased the errors on a_HVP(s) and on the intermediate window, so this approximation contributes to the final error budget, but no uncertainty is assigned to the scaling assumption itself. Please quantify the sensitivity of Eqs. (18)-(19) and Eqs. (22)-(23) to this assumption, for example by varying the proportionality factor within a conservative range or by computing ∂_c^sea on one ensemble with increased statistics, and add the corresponding systematic if it is not already covered.","section":"Appendix D, Eq. (D2)"}],"minor_comments":[{"comment":"The strange and charm analysis is stated to be unblinded; please include a brief statement of the analysis choices fixed in advance (e.g., model set, tmin and tcut criteria) so that the AIC-based error estimate can be assessed as a pre-specified procedure.","section":"Footnote 1"},{"comment":"There are several rendering or typographical artifacts, such as 'tcut7→∞' in Eq. (13) and 'Appedices' in the text of Section III; these should be corrected.","section":"Section III and Eq. (13)"},{"comment":"The many grey fit lines in the continuum-extrapolation plots are not individually labeled; adding a legend that maps each line to the corresponding AIC model variant would make the spread of the extrapolations easier to audit.","section":"Figs. 3, 4, and 6"},{"comment":"The comparison with other lattice groups neglects possible differences in the definition of isoQCD; a quantitative estimate of the expected size of these differences would help interpret the good agreement shown in Figs. 7 and 8.","section":"Footnote 5 and Figs. 7-8"},{"comment":"The AIC weight contains the term -2Ndata, which is nonstandard; since the authors state that the alternative definition from Ref. [38] gives similar results, reporting the shift in σ_cont between the two definitions would strengthen the robustness statement.","section":"Eq. (17)"}],"recommendation":"major_revision","confidential_remarks":"The paper is technically strong and the central values are likely correct, but the two major comments concern the dominant systematic uncertainty (continuum extrapolation) and an acknowledged approximation in the mistuning corrections. Both issues are addressable with additional analysis and reporting; if the requested leave-one-out and sensitivity checks confirm stability, I would support acceptance."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Solid, careful update of the ETMC strange and charm HVP determination. The new 0.049 fm ensemble and the explicit sea-quark and critical-mass mistuning corrections genuinely improve the error budget: the strange and charm connected contributions now come in at roughly 1.2% and 0.9% precision. This one deserves a serious referee.\n\nWhat is actually new: a fifth lattice spacing down to 0.049 fm, first-order reweighting corrections for sea-quark and critical-mass mistuning, and complete SD/W/LD window results for both flavors. The paper is transparent about its soft edges: the charm sea-quark derivative is estimated by a scaling ansatz (Eq. D2) rather than computed, the analysis is not blinded, and the gauge ensembles and code are not public. All three are stated openly, and none is a red flag at this precision.\n\nThe main strength is the error accounting. Statistical, continuum, and FSE errors are quoted separately (Eqs. 18-19). The continuum limit uses four lattice spacings with two regularizations (TM and OS) forced to a common limit, the AIC average spans a^2, a^4, and a^2 log^n forms plus cuts that drop the coarsest or next-to-coarsest spacing, and the tmin-branch analysis (Fig. 5) is a genuinely independent check on the short-distance region. The final numbers agree with every other lattice determination in Figs. 7-8, which is the external consistency check that matters.\n\nThe soft spot is the one the stress-test names: continuum-ansatz completeness. The cont error (0.48 for strange) is the largest single error, and with four spacings spanning a factor of about 2.7 in a^2, the data cannot sharply discriminate the functional forms. That is structural to the field rather than a flaw in execution, and the TM/OS joint fit does test the a^2 coefficient since the two regularizations carry different discretization artifacts. I would ask the authors for a leave-one-out fit that drops the new 0.049 fm point, to show the quoted cont error does not hinge on it alone. Minor.\n\nI mostly agree with the reader's take. The central claim holds up: the quark-connected strange and charm HVP contributions are known at these precisions in isoQCD. The citation pattern is clean; the comparison list covers BMW, CLS/Mainz, RBC/UKQCD, HPQCD, and the data-driven reviews. The audience is lattice QCD practitioners and the muon g-2 theory community, and they will get genuine value from the mistuning-correction methodology and the error budget.\n\nRecommendation: accept for peer review, with a referee who will check the AIC weighting and the mistuning derivatives.","headline":"A careful, transparent incremental update of the ETMC strange/charm HVP results; the new finer lattice spacing and explicit mistuning corrections are real improvements, and it deserves a serious referee.","tokens_in":36716,"tokens_out":5694,"would_cite":true,"duration_ms":44617,"reading_group":"yes","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":[],"pacs":["12.38.Gc","14.60.Ef"],"model":"deepseek-v4-flash","headline":"Lattice QCD sets the strange- and charm-quark connected HVP contributions to the muon's anomalous magnetic moment at (53.57 ± 0.63) and (14.56 ± 0.13) × 10^-10.","keywords":["muon g-2","hadronic vacuum polarization","lattice QCD","twisted-mass fermions","strange quark contribution","charm quark contribution","continuum extrapolation","AIC model averaging"],"falsifier":"A concrete test: compute the same strange and charm HVP contributions on an additional lattice spacing below 0.04 fm, or with an independent fermion action whose cutoff artifacts are known to differ, and see whether the continuum-extrapolated values remain within 53.57 ± 0.63 and 14.56 ± 0.13 (× $10^{-10}$); a shift exceeding the combined errors would show the model family is too narrow.","tokens_in":35436,"feed_emoji":"🧲","tokens_out":10682,"duration_ms":88214,"temperature":0.7,"pith_summary":"This paper aims to establish first-principles lattice QCD values for the strange- and charm-quark connected contributions to the hadronic vacuum polarization (HVP), the largest theoretical uncertainty in the muon anomalous magnetic moment. Using gauge ensembles with Nf = 2+1+1 flavors of Wilson-clover twisted-mass quarks at four lattice spacings, the authors correct small sea-quark and critical-mass mistunings to land on the isospin-symmetric physical point and then extrapolate to the continuum with an information-criterion model average. The claimed results are a_mu^HVP(s) = (53.57 ± 0.63) × $10^{-10}$ and a_mu^HVP(c) = (14.56 ± 0.13) × $10^{-10}$. These values matter because independent lattice determinations of the flavor-by-flavor HVP contributions are needed to test the tension between e+e- data-driven estimates and the experimental g-2; if correct, they show the strange and charm channels agree with other lattice groups and are not the origin of that tension.","feed_headline":"Strange and charm quark pieces of muon g-2 pinned to 1 percent","feed_subtitle":"First-principles values agree with other lattice groups and sharpen the hadronic vacuum polarization cross-check.","key_machinery":"The central object is the time-momentum representation of the HVP, which writes a_mu^HVP = 2 $alpha_em^{2}$ ∫_0^∞ dt $t^{2}$ K(m_mu t) V(t), where K is a known leptonic kernel and V(t) is the Euclidean vector-current correlator. On the lattice, V(t) is computed with the TM and OS regularizations, two discretizations that differ only by O($a^{2}$) artifacts, so forcing a common continuum limit in joint fits exposes the cutoff effects. The continuum extrapolation uses the ansatz P0 + P1 $a^{2}$ + P2 $a^{4}$ and variants with $a^{2}$/[log($a^{2}$/$lambda_0^{2}$)]^n terms, averaged with AIC weights; separately, leading-order reweighting converts the simulated ensembles to the isospin-symmetric point defined by the hadronic inputs M_pi = 135.0 MeV, M_K = 494.6 MeV, M_Ds = 1967 MeV, F_pi = 130.5 MeV.","core_discovery":"The central claim, stated as Eq. (1) and Eqs. (18)-(19), is that in isospin-symmetric QCD the quark-connected strange and charm HVP contributions are a_mu^HVP(s) = 53.57(63) × $10^{-10}$ and a_mu^HVP(c) = 14.56(13) × $10^{-10}$, where the error combines statistics, continuum extrapolation, finite-size, and tuning uncertainties in quadrature. The values come from vector correlators evaluated in two lattice regularizations that share the same continuum limit, corrected by leading-order reweighting for the small mistunings of the simulated bare parameters, and extrapolated with an AIC-weighted average over fit families of the form P0 + P1 $a^{2}$ + P2 $a^{4}$ (with logarithmic variants) using data at a = 0.049, 0.057, 0.068, and 0.080 fm. The same analysis yields short-distance, intermediate-window, and long-distance window contributions; the short-distance strange and intermediate-window charm determinations are the most precise lattice results in those windows.","pith_inferences":["Beyond the paper: if the light-quark connected and disconnected contributions from the same ensembles reach comparable precision, the combined lattice HVP prediction will provide a sharp, independent test of whether e+e- data or lattice QCD is closer to the experimental g-2 value.","Beyond the paper: the AIC error is a conditional estimate — it measures spread within the assumed a^2/a^4 ansatz family. A future lattice spacing below 0.04 fm, or a different action with different cutoff effects, is the natural check; a shift larger than the quoted error would indicate the fit family is incomplete.","Beyond the paper: the charm sea-quark derivative was approximated by a mass-scaling assumption in Appendix D; computing it directly would test whether the strange-quark result carries a small unquantified bias from that approximation."],"forward_implications":["The strange and charm connected HVP contributions are now established at sub-percent precision, so future full lattice determinations of a_mu^HVP can treat these channels as fixed anchors rather than as free sources of error.","The short-distance strange value, 9.063(27) × 10^-10, and intermediate-window charm value, 2.920(64) × 10^-10, are the most precise lattice results in their respective windows, sharpening window-by-window comparisons with e+e- data.","Agreement with the other lattice determinations shown in the paper indicates that the longstanding g-2 tension is not driven by the strange or charm connected contributions.","The new fine lattice spacing (a ≈ 0.049 fm) and the explicit mistuning corrections reduce the total uncertainty relative to the earlier determination of Ref. [22], making these the reference values for those channels until the light-quark and disconnected contributions are added."],"supporting_citations":[{"why":"Supplies the window definitions and the hadronic method for the ZV and ZA renormalization constants; also the previous determination that this calculation improves.","marker":"[22]"},{"why":"Describes the generation and tuning of the Nf = 2+1+1 Wilson-clover twisted-mass gauge ensembles used here.","marker":"[23]"},{"why":"Defines the target isospin-symmetric physical point (M_pi, M_K, M_Ds, F_pi) that the tuning procedure matches.","marker":"[28]"},{"why":"Introduces the time-momentum representation and the kernel K(m_mu t) that converts the vector correlator into a_mu^HVP.","marker":"[32]"},{"why":"Provides the NNLO perturbative input used to add the [0, t_min] region in the second analysis branch.","marker":"[35]"},{"why":"Supplies the AIC weighting and error-combination procedure used for the continuum-limit model average.","marker":"[36]"},{"why":"Introduces the short-distance, intermediate-window, and long-distance window decomposition used for the partial contributions.","marker":"[21]"},{"why":"Defines the AIC criterion whose weights select the continuum extrapolation models.","marker":"[38]"}],"fun_headline_variants":["Lattice QCD pins down strange and charm g-2 pieces","Muon g-2: lattice sharpens strange and charm quark terms","Strange and charm quark g-2 contributions from lattice QCD","Lattice QCD improves precision of strange and charm g-2","Muon g-2: strange and charm contributions pinned"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The load-bearing premise is that the continuum-limit fit family — smooth $a^{2}$ and $a^{4}$ terms with optional logarithmic variants — spans the true discretization effects at the four lattice spacings used.","fun_headline_variants_meta":{"raw":{"variants":["Lattice QCD pins down strange and charm g-2 pieces","Muon g-2: lattice sharpens strange and charm quark terms","Strange and charm quark g-2 contributions from lattice QCD","Lattice QCD improves precision of strange and charm g-2","Muon g-2: strange and charm contributions pinned"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.001172,"raw_usage":{"total_tokens":4891,"prompt_tokens":1035,"completion_tokens":3856,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":651,"completion_tokens_details":{"reasoning_tokens":3766}},"tokens_in":651,"tokens_out":3856,"duration_ms":22273,"temperature":1.0,"reasoning_tokens":3766,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-12T21:16:07.236018+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"A concrete test: compute the same strange and charm HVP contributions on an additional lattice spacing below 0.04 fm, or with an independent fermion action whose cutoff artifacts are known to differ, and see whether the continuum-extrapolated values remain within 53.57 ± 0.63 and 14.56 ± 0.13 (× $10^{-10}$); a shift exceeding the combined errors would show the model family is too narrow.","supporting_citations":[{"cited_title":"Akaike, A new look at the statistical model identification , IEEE Transactions on Automatic Control 19 (1974) 716","cited_arxiv_id":null,"evidence_quote":"Defines the AIC criterion whose weights select the continuum extrapolation models."}],"review_version":1}