{"id":"38bd1c83-f068-482a-9d42-307b1ef68eff","arxiv_id":"2502.06305","paper_version":1,"verdict":"CONDITIONAL","confidence":"HIGH","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":6,"one_line_summary":"This lattice QCD proceedings paper presents a stable analysis strategy and preliminary results for the isoscalar non-singlet axial form factor of the nucleon, with the full u+d-2s case on two ensembles.","lead":"Scientists used lattice QCD, a supercomputer-based method, to compute a property of the nucleon called the axial form factor, focusing on the quark combination that isolates strange quark effects. The paper is a progress report: the analysis strategy is shown to be stable, and preliminary results agree with earlier model estimates.","discovery_kind":"extension","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The summation-method window average does not by itself rule out residual excited-state contamination; a direct two-state or higher-ts comparison would settle whether the slope in Eq. (7) is clean.","rationale":"The paper is a careful proceedings contribution that clearly labels its results as preliminary, and it provides useful cross-checks: multiple covariance regularisations agree, the b0 parametrisation is varied, and the connected-data chiral-continuum extrapolation is handled with several ansaetze and an AIC model average. These are real strengths. However, the central methodological claim is that the combination of the summation method, direct z-expansion, and window average yields a stable route to G_A^{u+d-2s}(Q^2). That claim rests on the slope of Eq. (7) being dominated by the ground state over the fitted ts range. The window average over ts,min is not an independent excited-state test: it reweights fits that all include ts values in the same 0.8-1.0 fm region, and the O(ts e^{-Delta ts}) term in Eq. (7) is not estimated or bounded anywhere in the paper. Since the quoted agreement with external values uses exactly this slope, residual contamination would directly invalidate the main quantitative comparison. This is the same weakest assumption identified by the reader, and it is the most load-bearing single point: if it fails, the extracted form factor and its Q^2 dependence are biased. A concrete two-state or variable-ts-min test would settle the matter. The appropriate verdict remains CONDITIONAL, because the paper is explicitly preliminary and the missing check is a natural next step rather than a demonstrated internal inconsistency.","tokens_in":8404,"tokens_out":4257,"duration_ms":39714,"concrete_test":"On E300 (or D200), repeat the direct z-expansion fit with ts,min = 0.8, 1.0, 1.2 fm and extract G_A at a fixed Q^2 (e.g. Q^2 = 0.2 GeV^2) for each choice; then perform a correlated fit including an excited-state term A e^{-Delta ts} with Delta ~ m_N* - m_N as a nuisance parameter. If the slope changes by more than one combined standard deviation relative to the window-averaged value, Eq. (7)'s O(ts e^{-Delta ts}) term is significant and the current estimate is biased.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The most load-bearing assumption is that the slope of the summed ratio S(q,ts) in Eq. (7) is dominated by the ground-state form factor for the ts values entering the window average. The window in Eq. (9) concentrates weight on ts,min between 0.8 and 1.0 fm; it does not test convergence of the linear-in-ts behaviour. The fits for each ts,min all include ts values at or above that minimum, so if excited states with e^{-Delta ts} are not negligible at ts ~ 1 fm, the extracted slope G_A(Q^2) is biased. The paper demonstrates stability under covariance regularisation and under the b0 parametrisation, but not under a variation of the excited-state model: no two-state fit or explicit O(ts e^{-Delta ts}) check is reported. Because the final comparison to g_A^{u+d-2s}=0.46(5) and 0.490(20) uses exactly this slope, a positive excited-state contamination would make the apparent agreement coincidental. This is a correctness risk, not a disagreement with consensus.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"This proceedings paper reports work in progress on the isoscalar non-singlet axial form factor G_A^{u+d-2s}(Q^2) of the nucleon from N_f=2+1 CLS lattices. The analysis combines the summation method for excited-state suppression with a direct z-expansion fit at order n=2, a window average over the minimum source-sink separation, two covariance-matrix regularisations, and an AIC model average for the chiral-continuum extrapolation. The connected u+d channel is carried through the full extrapolation, while the full u+d-2s result is shown on two ensembles only (E300 and D200). The authors report compatibility of the preliminary octet axial charge with the Cloudy Bag value g_A^{u+d-2s}=0.46(5) and with the ETM value 0.490(20).","tokens_in":8659,"tokens_out":5392,"duration_ms":52576,"significance":"If completed, this study would provide the first lattice QCD determination of the isoscalar octet axial form factor in a range relevant for MicroBooNE and for flavour decompositions of the nucleon spin. The paper's strengths are its transparent internal checks: the direct z-fit is compared with the two-step procedure, the b0(ts) intercept is treated both freely and via a z-expansion, off-diagonal damping and SVD cuts are compared with the unregulated covariance matrix, and the model average uses correlated coefficients. The use of independent renormalisation factors and the explicit statement of remaining steps are also commendable. The main weaknesses are the lack of an explicit excited-state contamination test and the absence of a continuum extrapolation for the full octet channel, both of which are acknowledged in spirit but not quantified.","major_comments":[{"comment":"The central extraction assumes that, for the t_s values entering the window average, the O(t_s e^{-Delta t_s}) term in Eq. (7) is negligible, so that G_A(Q^2) is the slope of S(q,t_s). The window average over t_s,min does not test this assumption: every fit still uses all t_s >= t_s,min, so an excited-state contamination that is non-negligible at t_s ~ 1 fm would bias all fits in the same direction. The paper validates stability under covariance-matrix regularisation and under different parametrisations of b0, but reports no two-state fit and no explicit study of the exponential term. Since the final comparison to g_A^{u+d-2s}=0.46(5) and 0.490(20) uses exactly this slope, a positive contamination would make the agreement coincidental. Please add a direct excited-state check (for example a two-state fit or a comparison at substantially larger t_s) or state explicitly that this systematic is not yet controlled.","section":"§3, Eqs. (7)–(9), Fig. 2"},{"comment":"The full u+d-2s result is presented on two ensembles, E300 and D200, without a continuum extrapolation. The text does label the evaluation as preliminary and notes that lattice artifacts are not accounted for, but the subsequent sentence, 'these provide a value of the axial charge compatible with...', presents the comparison as validation. Because the two ensembles may share common discretisation effects, agreement with [19,20] is weaker evidence than a single-ensemble comparison would be. Please either soften the claim to 'consistent within statistical errors, with missing systematics not yet included' or provide a quantitative estimate of the size of the omitted discretisation, finite-volume, and renormalisation uncertainties.","section":"§4, Fig. 5 and final paragraph"}],"minor_comments":[{"comment":"The sentence beginning 'The axial form factors G_A(Q^2) of the nucleon plays' has a subject-verb agreement error; it should be 'play' or the sentence should be restructured with a singular subject.","section":"§1, first sentence"},{"comment":"The red curve is described as a 'zoom on the window function', but it is plotted with the z-expansion coefficients on the same axes without specifying its scale or normalisation; this is confusing and should be clarified.","section":"Fig. 2 caption"},{"comment":"The z-expansion is fixed at order n=2 for the central coefficients and only the order n_b for b0 is varied in Fig. 1. Since the z-expansion truncation is a systematic assumption, the authors should state whether an order n=3 fit or an alternative parametrisation has been tried or is planned beyond the dipole mentioned in the outlook.","section":"§3, after Eq. (8)"},{"comment":"The comparison with g_A^{u+d-2s}=0.46(5) and 0.490(20) is made without showing the E300 and D200 values or their uncertainties in the text or a table; providing these numbers would make the level of agreement quantitative.","section":"§4, final paragraph"},{"comment":"The sentence 'we take the factors Z_A from [9] and b_A from [10], neglecting the coefficient \\tilde b_A and f_A' would benefit from a brief justification of why those coefficients are expected to be small in this channel, since the neglected terms are part of the O(a) improvement of the axial current.","section":"§2, renormalisation paragraph"}],"recommendation":"major_revision","confidential_remarks":"This is a proceedings contribution with explicitly preliminary results, and the authors are transparent about the missing steps. My main concern is the untested excited-state assumption in the summation method, which is load-bearing for the quoted compatibility with external values. If the authors add a short discussion of the excited-state systematic or a concrete test, I would be willing to support acceptance. The comparison with g_A values should be framed as an intermediate cross-check rather than a result."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Here's my read. This is a solid proceedings paper from a group that knows what it's doing. The genuinely new element is the one-step direct z-fit for the u+d-2s axial form factor: instead of extracting G_A per Q^2 and then fitting z, they fit the z-coefficients directly to the summed-ratio data for all ts and Q^2 simultaneously. They validate this against the two-step approach and show they agree. The window average over ts_min, the AIC model averaging, and the covariance regularization checks (correlated, off-diagonal damping, SVD) are all sensible and reported cleanly. The results are flagged as preliminary, and the outlook honestly lists what's missing: disconnected contributions on all ensembles, continuum extrapolation of the full octet case, more fit ansätze, and a complete error budget.\n\nThe main soft spot is exactly what you'd expect from a progress report: excited-state control. The summation method assumes the slope in Eq. (7) is linear in ts with negligible O(ts e^{-Delta ts}) contamination. The window average over ts_min is a reasonable way to reduce human bias, and the plateau in Fig. 2 is reassuring, but it is not a substitute for a two-state fit or an explicit check at larger ts. The stress-test note is right that positive excited-state contamination would bias all the G_A values, and the apparent agreement with external g_A values could be coincidental. I don't think this is a load-bearing flaw in a paper that explicitly says it is preliminary, but the final paper should close this gap.\n\nThe full octet result currently rests on two ensembles with no continuum extrapolation, and the z-order is fixed at n=2. These are acknowledged limitations, not hidden ones. The citation pattern is clean: they rely on their own previous work [15] for the established strategy, and take renormalization factors from independent groups.\n\nBottom line: this is a legitimate, careful contribution for the lattice/neutrino community. It doesn't overclaim. I'd send it to a serious referee if it were a journal submission; for a proceedings, it's above average. The excited-state question is the thing to watch in the full paper.","headline":"A careful, honest progress report on the isoscalar octet axial form factor; the one-step z-fit is well validated, and the remaining excited-state and continuum systematics are explicitly deferred to the final paper.","tokens_in":9238,"tokens_out":4101,"would_cite":false,"duration_ms":37437,"reading_group":"maybe","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":[],"pacs":["12.38.Gc"],"model":"deepseek-v4-flash","headline":"The paper shows that combining the summation method with a direct z-expansion fit, a window average over source-sink separations, and an AIC model average gives a stable lattice extraction of the nucleon's isoscalar non-singlet axial form…","keywords":["axial form factor","isoscalar non-singlet","lattice QCD","nucleon axial charge","summation method","z-expansion","model averaging","strange axial form factor"],"falsifier":"Compute the summed ratio on a single ensemble for at least four source-sink separations above 1 fm and check whether the slope in $t_s$ is independent of the fitted $t_s$ window and whether the residuals show the expected $O(t_s e^{-\\Delta t_s})$ curvature; if the slope shifts with the window, the window-averaged extraction is biased.","tokens_in":8176,"feed_emoji":"⚛️","tokens_out":9500,"duration_ms":80647,"temperature":0.7,"pith_summary":"This proceedings paper argues that a specific analysis pipeline extracts the isoscalar non-singlet axial form factor of the nucleon, $G_A^{u+d-2s}(Q^2)$, from lattice QCD in a controlled way up to momentum transfers of $0.7$ GeV$^2$. The pipeline combines the summation method for suppressing excited-state contamination with a direct $z$-expansion fit over all $Q^2$ and all source-sink separations, a window average over the minimum separation, and an AIC model average over chiral-continuum ansaetze. This matters because this flavour combination, together with the singlet channel, determines the strange axial form factor that neutrino-scattering experiments are trying to measure. On two of the most chiral ensembles the preliminary octet axial charge is compatible with the established values $g_A^{u+d-2s}=0.46(5)$ and $0.490(20)$, and the connected contribution already follows the extrapolated physical curve.","feed_headline":"Lattice QCD zeroes in on nucleon's isoscalar axial form factor","feed_subtitle":"New analysis chain keeps excited states in check and matches established octet axial-charge values.","key_machinery":"The carrying mechanism is the summation-method identity $S(\\boldsymbol{q},t_s)=b_0(Q^2)+t_s\\,G_A(Q^2)+O(t_s e^{-\\Delta t_s})$, which turns the form factor into the slope of a linear fit in the source-sink separation $t_s$. The paper combines it with a $z$-expansion in the variable $z(Q^2)=\\big(\\sqrt{t_{\\rm cut}+Q^2}-\\sqrt{t_{\\rm cut}}\\big)/\\big(\\sqrt{t_{\\rm cut}+Q^2}+\\sqrt{t_{\\rm cut}}\\big)$ truncated at order $n=2$, fits all $Q^2\\le 0.7$ GeV$^2$ and all $t_s$ in one go, and then averages over $t_{s,\\min}$ with a tanh window (lower edge $0.8$ fm, upper edge $1$ fm, width $0.08$ fm). The $z$-expansion absorbs the $Q^2$ dependence, the window average removes the human choice of $t_{s,\\min}$, and the later AIC model average combines chiral-continuum ansaetze and cuts. The large covariance matrix is regulated either by damping the off-diagonal elements or by an SVD cut.","core_discovery":"The central claim is that the direct $z$-fit route is a valid replacement for the traditional two-step procedure: instead of extracting $G_A(Q^2)$ point by point and then fitting $z$, one fits the expansion coefficients $a_0,a_1,a_2$ in a single simultaneous fit over all $Q^2\\le 0.7$ GeV$^2$ and all source-sink separations, treating the constant $b_0(Q^2)$ of the summation method either as a free parameter per $Q^2$ or as a second $z$-expansion; both give compatible results. The paper demonstrates that covariance-matrix regularization by off-diagonal damping or by an SVD cut does not change the outcome, and that a window average over $t_{s,\\min}$ with fixed physical windows removes the main human bias in choosing the fit range. On the two most chiral ensembles the full $u+d-2s$ form factor, with disconnected loops included, shows a low-$Q^2$ shift relative to the connected contribution, and the axial charge agrees with existing benchmarks. The paper presents this as progress toward a first physical result for the isoscalar octet form factor, with the same strategy to be applied to the full ensemble set.","pith_inferences":["If the linear-dominance assumption in the summation method holds at $t_s\\ge0.8$ fm, the same window-averaged direct-fit strategy should suppress excited states just as well for other baryon form factors, such as the electromagnetic or induced-pseudoscalar ones.","The fixed choices $t_{\\rm cut}=(4M_\\pi)^2$ and $Q^2_{\\max}=0.7$ GeV$^2$ could be tested by extending the fit to $1$ GeV$^2$; a failure of the $n=2$ truncation there would reveal whether the low-$Q^2$ stability is a property of the method or of the chosen range.","A sharper test of the method would be to compare results from the window-averaged summation fit with a variational multi-operator estimate of the first excited-state energy and amplitude on the same ensembles.","The same framework could be applied to the singlet channel, turning the flavour decomposition of the axial form factor into one unified fit rather than separate connected and disconnected analyses."],"forward_implications":["The same analysis chain can be run on the full ensemble set to produce the first physical result for the isoscalar octet axial form factor over $Q^2\\in[0,0.7]$ GeV$^2$.","Combined with an analogous singlet $u+d+s$ computation, the method opens a route to the strange axial form factor $G_s^A(Q^2)$ in the range relevant to neutrino-nucleus experiments.","Because the window average and the model average are applied with fixed choices, the quoted errors include a systematic spread from analysis choices rather than one hand-picked fit.","The agreement of the preliminary axial charge with external estimates at low $Q^2$ supports the treatment of disconnected contributions and excited states at the current statistical precision.","The connected $u+d$ result already lies close to the physical curve after the chiral-continuum extrapolation, suggesting the remaining corrections on the most chiral ensembles are small."],"supporting_citations":[{"why":"Supplies the analysis strategy this work extends, including the direct $z$-fit and off-diagonal covariance damping.","marker":"[15]"},{"why":"Introduces the summation method that produces the linear-in-$t_s$ identity used to extract the form factor.","marker":"[16]"},{"why":"Demonstrates the summation-method extraction of the nucleon axial charge with controlled errors.","marker":"[17]"},{"why":"Supplies the AIC-based model averaging used to combine chiral-continuum ansaetze and cuts.","marker":"[18]"},{"why":"Provides the phenomenological octet axial charge $0.46(5)$ used as a benchmark.","marker":"[19]"},{"why":"Provides the recent lattice result $0.490(20)$ used as a benchmark for the octet axial charge.","marker":"[20]"},{"why":"Provides the $N_f=2+1$ ensembles with improved fermions on which all results are based.","marker":"[11]"}],"fun_headline_variants":["Single z-fit rivals two-step for nucleon axial form factor","Lattice QCD streamlines isoscalar axial form factor extraction","Direct z-expansion fit validated for nucleon axial form factor","One-shot fit matches traditional method for nucleon's axial form","New fit straight to z-coefficients for nucleon axial form factor"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The load-bearing premise is that the summation-method identity is dominated by its linear term for the chosen source-sink separations, so that excited-state contamination of order $t_s e^{-\\Delta t_s}$ is negligible after window averaging.","fun_headline_variants_meta":{"raw":{"variants":["Single z-fit rivals two-step for nucleon axial form factor","Lattice QCD streamlines isoscalar axial form factor extraction","Direct z-expansion fit validated for nucleon axial form factor","One-shot fit matches traditional method for nucleon's axial form","New fit straight to z-coefficients for nucleon axial form factor"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000236,"raw_usage":{"total_tokens":1536,"prompt_tokens":1013,"completion_tokens":523,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":629,"completion_tokens_details":{"reasoning_tokens":433}},"tokens_in":629,"tokens_out":523,"duration_ms":4712,"temperature":1.0,"reasoning_tokens":433,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-08T15:51:57.410698+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Compute the summed ratio on a single ensemble for at least four source-sink separations above 1 fm and check whether the slope in $t_s$ is independent of the fitted $t_s$ window and whether the residuals show the expected $O(t_s e^{-\\Delta t_s})$ curvature; if the slope shifts with the window, the window-averaged extraction is biased.","supporting_citations":[{"cited_title":"Maiani, G","cited_arxiv_id":null,"evidence_quote":"Introduces the summation method that produces the linear-in-$t_s$ identity used to extract the form factor."},{"cited_title":"The nucleon's octet axial-charge g_A^8 with chiral corrections","cited_arxiv_id":"0912.1765","evidence_quote":"Provides the phenomenological octet axial charge $0.46(5)$ used as a benchmark."}],"review_version":1}