{"id":"466adeb7-82f0-4c95-ae3b-e73e5f5b18d0","arxiv_id":"2505.05941","paper_version":2,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":9,"one_line_summary":"The complete order-M^4 two-loop corrections to g_A are derived in EOMS and infrared covariant baryon chiral perturbation theory and evaluated with scattering-derived low-energy constants.","lead":"This paper computes the two-loop corrections to the nucleon's axial-vector coupling constant g_A in two versions of covariant chiral perturbation theory. The new order-M^4 terms are small for one set of low-energy constants and moderate for another, which matters for lattice QCD extrapolations and tests of the chiral expansion.","discovery_kind":"extension","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Numerical results mix HB and EOMS LECs: the paper identifies HB bar d_i with EOMS d_i^r without applying the finite scheme shifts that Eq. (5) acknowledges for g0 and d16.","rationale":"The reader's verdict (CONDITIONAL) and weakest assumption (uncomputed higher-order terms) are reasonable, and I agree that the numerical conclusion is not fully auditable. However, the most load-bearing concern for the central claim is not the acknowledged M^5/q^6 pieces but an apparent scheme mismatch in the input LECs. The paper's numerical results rely on fitting or equating HB LECs with EOMS renormalized LECs without showing the finite shifts that Eq. (5) demonstrates are necessary for g0 and d16. Since Eqs. (32) and (34) are explicitly written in terms of d_i^r(mu), plugging in bar d_i^{HB} values is only justified if the matching is provided; the paper's statement that the expanded EOMS results 'correspond' to HB is an assertion, not a derivation. This concern is concrete and testable, and it directly affects the quoted percentages rather than merely the convergence interpretation. I also note the F_pi = 0.927 GeV typo and the unstated g0 fixing procedure as secondary but real auditability issues. Because these issues all point to the same conditional status rather than to a demonstrated error, I recommend keeping the reader's CONDITIONAL verdict (UNCHANGED).","tokens_in":11875,"tokens_out":12710,"duration_ms":132440,"concrete_test":"Recompute Eq. (35) for both LEC sets using the same two-loop formulas but with the HB LECs of Ref. [36] converted to the EOMS scheme via the finite matching relations for d_i and l_i (the analogue of the g0 and d16 shifts in Eq. (5), using Ref. [18] for the EOMS delta_i). If either Delta(4) value changes by more than about 5 percentage points - for example, if set 2's -17.4% becomes positive or changes in magnitude by a factor of ~1.5 - the numerical central claim is not scheme-consistent as presented and the conclusion of 'moderate' corrections is not robust.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central quantitative claim is that the leading two-loop corrections are moderate, with Delta(4) = 2.6% (set 1) and -17.4% (set 2) at the physical pion mass. These numbers are obtained by inserting LECs 'from an analysis of elastic and inelastic pion-nucleon scattering [36]' into the EOMS expressions (32) and (34). The manuscript states that these central values 'have been obtained in heavy baryon chiral perturbation theory' and then asserts that 'the bar d_i are the d_i^r(mu) defined at mu=M_pi' and that 'as we have done an expansion in M/m of the IR and EOMS expressions our results should in fact correspond to the one in the heavy baryon approach.' This identification is a load-bearing assumption, not a demonstrated equivalence: EOMS and HB renormalized LECs differ by finite scheme-dependent shifts (explicitly displayed for g0 and d16 in Eq. (5)), and the same type of shifts generically affect the d_i and l_i appearing in Eqs. (32) and (34). If the conversion between bar d_i^{HB} and d_i^{EOMS} is not applied, the numerical values of gamma4 and beta4 - and hence the quoted percentages, which are evaluated at C0=0 - are scheme-inconsistent. Because set 2's Delta(4) is already -17.4%, a shift of only a few percent from a proper LEC matching would change the 'moderate' conclusion substantially. The factor-of-10 typo in F_pi = 0.927 GeV and the absence of an uncertainty analysis further obscure the numerical section, but the LEC scheme matching is the more fundamental issue for the numerical central claim.","agreement_with_reader":"partial"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper computes the leading two-loop (chiral order q^5) corrections to the nucleon axial-vector coupling g_A in two covariant formulations of two-flavor baryon chiral perturbation theory, EOMS and infrared regularization. The central analytic results are the coefficients alpha4 in Eq. (30), gamma4 in Eq. (32), and beta4 in Eq. (34), together with the claim that the two-loop M^4 contribution to g_A is moderate once existing pion-nucleon scattering LECs are inserted: Delta(4) = 2.6% for set 1 and -17.4% for set 2 at the physical pion mass. The derivation is supported by several internal checks, including cancellation of log/epsilon divergences, scale independence of the counterterm combination C, agreement with known two-loop nucleon self-energy results, and agreement with the g0^5 term of Ref. [14] after correcting a factor of two. The numerical part uses two sets of LECs from heavy-baryon analyses of pion-nucleon scattering and assumes the order-q^5 counterterm C0 to be of natural size.","tokens_in":12361,"tokens_out":8552,"duration_ms":90533,"significance":"If the analytic coefficients are correct, this is the first complete covariant two-loop M^4 calculation of g_A in EOMS and infrared regularization, and the result for alpha4 in Eq. (30) is a parameter-free prediction in terms of g0. The consistency checks cited in the paper strengthen confidence in the reduction. The numerical claim of moderate two-loop corrections is, however, less firmly established: it depends on uncalculated higher-order pieces, on an LEC scheme identification that is asserted rather than derived, and on a numerical input set whose reported central values are not accompanied by uncertainties. The analytic part is likely to be of lasting value; the convergence conclusion needs additional support or more careful wording.","major_comments":[{"comment":"The quoted Delta(4) values are obtained by inserting the heavy-baryon bar d_i LECs of Ref. [36] directly into the EOMS expressions (32) and (34). This is a load-bearing scheme identification: Eq. (5) explicitly shows that at least g0 and d16 require finite scheme-dependent shifts between EOMS and other schemes, and analogous shifts generically affect the other d_i and l_i appearing in gamma4 and beta4. The statement in Sec. 4 that 'as we have done an expansion in M/m of the IR and EOMS expressions our results should in fact correspond to the one in the heavy baryon approach' is not a derivation. Because the percentages in Eq. (35) are evaluated at C0=0, any finite shift in gamma4 or beta4 changes the quoted numbers. Please either apply the explicit HB-to-EOMS conversion for all LECs used, or demonstrate that the conversion only renormalizes C0 and that the resulting variation is within the C0 = +/-15 GeV^-4 band already shown in Fig. 3.","section":"Sec. 4, Eqs. (32)-(35)"},{"comment":"The paper states that 'we find that g0 = 1.0 for set 1 and g0 = 1.3 for set 2, in order,' but it does not explain how g0 is determined. The quoted Delta(n) values imply that g0 is chosen so that the truncated series reproduces the physical g_A; this should be stated explicitly, since otherwise the smallness of Delta(4) at the physical point is partially an artifact of the fitting condition. More importantly, the convergence interpretation is not yet supported: for set 2, Delta(3) = 44.5% and Delta(4) = -17.4% are of comparable magnitude, and the paper itself notes in Secs. 2 and 5 that the order-q^6 two-loop diagrams and M^5 corrections are needed before final conclusions can be drawn. The abstract's claim that 'these corrections are rather moderate' should be restricted to the computed Delta(4) contribution, with the caveat that the omitted higher-order pieces have not been shown to be small.","section":"Sec. 4, Eq. (35) and Fig. 3"},{"comment":"The manuscript defers all details of the 44-diagram two-loop calculation to Ref. [20], which is listed as 'in preparation.' Since Eqs. (30), (32), and (34) are the central results of the paper, the reader cannot independently verify the reduction, the master-integral decompositions, or the LEC contributions. The internal consistency checks are reassuring, but they are not a substitute for a complete derivation. Please include the full calculation, or at least the master-integral reduction and the contribution of each class of diagrams/LECs, either in the paper or in a supplementary file that is made available with the submission.","section":"Sec. 3, around Eqs. (12)-(16) and (30)-(34)"},{"comment":"The numerical percentages are quoted without any uncertainty estimate, despite the dependence on LEC fit errors, the unconstrained choice C0=0, and the scheme-matching issue raised above. In particular, for set 2 a shift of Delta(4) by a few percent would change the conclusion from 'moderate' to 'substantial.' Given that the full uncertainty analysis is deferred to Ref. [20], the quantitative claims in Eq. (35) should either be accompanied by an error estimate or be explicitly labeled as central values that are indicative only. At minimum, the corresponding Delta(4) values for C0 = +/-15 GeV^-4 should be listed, since Fig. 3 shows that the M^4 curves vary appreciably over that range.","section":"Sec. 4, Eq. (35)"}],"minor_comments":[{"comment":"The value 'F_pi = 0.927 GeV' is a factor of 10 too large; the physical value is 0.0927 GeV. If the numerical evaluation actually used F = 0.0927 GeV (as the quoted results suggest), the text should be corrected to avoid making the calculation irreproducible.","section":"Sec. 4"},{"comment":"The sentence giving set 2 contains a typographical error: 'bar d_1 + bar d_2 = 3.39,GeV^-2' and 'bar d_13 = 27.7,GeV^-2' have misplaced commas. Also, 'g0 = 1.0 for set 1 and 1.3 for set 2, in order' should read 'respectively.'","section":"Sec. 4"},{"comment":"The notation 'g0 -> g0 + delta g0|reg d16 -> d16 + delta d16|reg' is missing a separator; the two scheme shifts should be displayed as separate equations or separated by a comma for clarity.","section":"Eq. (5)"},{"comment":"There are minor proofreading issues, including the duplicated 'by by' in the acknowledgements and the misspelling 'Reserch' in the affiliation of the fourth author; these should be corrected.","section":"Acknowledgements and author affiliations"}],"recommendation":"major_revision","confidential_remarks":"The analytic calculation appears significant and is backed by several nontrivial internal checks, so I would not reject on the current evidence. The main risk is the numerical interpretation: the scheme identification of the LECs, the determination of g0, and the absence of uncertainty estimates together make the 'moderate corrections' claim provisional. These issues are addressable within the manuscript's scope, either by adding the HB-to-EOMS conversion and an uncertainty estimate, or by explicitly softening the convergence claims pending the future calculation of the q^6 and M^5 contributions."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Colleague,\n\nThe short version: this is a genuine two-loop calculation, the first complete covariant M^4 result for g_A in both EOMS and IR schemes, and the analytic coefficients are backed by several honest checks. But the abstract's \"rather moderate\" convergence claim is conditional, and the weakest joint is numerical: the LEC inputs are taken from heavy-baryon analyses and inserted into EOMS expressions without demonstrating the required finite scheme shifts.\n\nWhat's actually new and good: alpha4, gamma4, beta4 at order M^4 in both EOMS and IR; the two-loop wave-function renormalization; and the corrected factor-of-two error in Ref. [14]. The paper does not just quote results. It checks the cancellation of log/epsilon divergences, the scale independence of the q^5 counterterm C, agreement with published two-loop self-energy calculations, and the overlap with the g0^5 term of the HB/RG result. Those checks are meaningful. This paper deserves a serious referee.\n\nSoft spots, in proportion:\n\n1. The derivation is not actually in the paper. Everything is deferred to Ref. [20], \"in preparation.\" The final formulas are plausible, but no master-integral reductions or intermediate expressions are shown. The internal checks help, but a referee cannot audit the central result from this manuscript alone.\n\n2. The scheme identification for the LECs is asserted, not demonstrated. The text says the bar d_i from the HB fit are the d_i^r(mu) at mu = M_pi, and that the M/m expansion means the IR/EOMS result corresponds to HB. That is not the same as applying the EOMS-to-HB finite shifts, which the paper explicitly displays for g0 and d16 but does not implement for the other d_i and l_i. For set 2, where Delta(4) = -17.4%, a few percent shift in these LECs can change the numerical conclusion from \"moderate\" to \"large.\" This is the main reason I do not take the central quantitative claim at face value yet.\n\n3. F_pi = 0.927 GeV is a factor-of-10 typo. It does not appear to affect the quoted percentages, but it is a reproducibility red flag.\n\n4. There is no uncertainty on the percentages, the procedure for g0 (1.0/1.3, \"in order\") is not stated, and the paper itself says M^5 corrections are needed before drawing conclusions.\n\nBottom line: the analytic part deserves a serious referee and likely publication. The numerical part is not yet auditable. I would engage with the paper, but I would not cite the percentages until the LEC scheme matching is done properly and the full derivation is available.","headline":"Genuine first complete covariant two-loop g_A coefficients with plausible analytics, but the numerical 'moderate' claim rests on an unshown HB-to-EOMS LEC matching and on details deferred to a follow-up paper.","tokens_in":12880,"tokens_out":5064,"would_cite":true,"duration_ms":53421,"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":"The paper computes the leading two-loop corrections to the nucleon axial-vector coupling constant $g_A$ at chiral order $q^5$ in two covariant versions of two-flavor baryon chiral perturbation theory (EOMS and infrared regularization) and…","keywords":["axial-vector coupling constant","g_A","chiral perturbation theory","two-loop corrections","covariant baryon ChPT","EOMS scheme","infrared regularization","pion-nucleon scattering"],"falsifier":"Compute the order-$q^6$ two-loop diagrams with $L_{\\pi N}^{(2)}$ vertices and the $M^5$ corrections to $g_A$: if their combined contribution at the physical pion mass is comparable to or larger than the reported $|\\Delta^{(4)}|$ (in particular for LEC set 2, where $\\Delta^{(4)}=-17.4\\%$), the claim that the leading two-loop corrections are moderate is falsified. A complementary check is a high-precision lattice value of $g_A$ at the physical pion mass that cannot be accommodated once these higher-order terms are included at their natural size.","tokens_in":11675,"feed_emoji":"⚛️","tokens_out":10177,"duration_ms":90873,"temperature":0.7,"pith_summary":"This paper establishes the complete leading two-loop contribution to the nucleon axial-vector coupling $g_A$ at chiral order $q^5$, i.e. the $M^4$ term in the chiral expansion, working in the two covariant renormalization schemes of baryon chiral perturbation theory: extended-on-mass-shell (EOMS) and infrared regularization (IR). The centerpiece is a closed expression for the scheme-independent coefficient $\\alpha_4 = -\\frac{7}{3} g_0 (1-g_0^2) + 16 g_0^5$, along with the scheme-dependent coefficients $\\gamma_4$ and $\\beta_4$ given in Eqs. (32) and (34). Evaluated at physical pion mass with two published sets of pion-nucleon scattering low-energy constants, the two-loop correction is $\\Delta^{(4)} = 2.6\\%$ for one set and $-17.4\\%$ for the other, which the authors describe as 'rather moderate'. If correct, these coefficients are the first complete covariant two-loop $M^4$ results for $g_A$, and they give a controlled basis for chiral extrapolations of lattice QCD calculations of the axial coupling.","feed_headline":"gA two-loop correction: 2.6% to -17.4%","feed_subtitle":"The first complete covariant two-loop M^4 coefficients for gA are derived in EOMS and infrared schemes.","key_machinery":"The engine is the set of scalar two-loop master integrals $F_{\\alpha\\beta\\gamma\\delta\\epsilon}(m_1,\\ldots,m_5)$ with two integration momenta and up to five propagators, evaluated on shell ($p^2=m^2$) and expanded through order $M^5$. Each integral is split into a purely infrared part, a mixed part, and a regular part following the infrared-regularization decomposition; the regular parts are absorbed into low-energy constants, while the mixed and infrared parts satisfy the Ward identities. The wave-function renormalization factor $Z$ is built from the same integrals at one and two loops, and the requirement that the $\\log(M/\\mu)/\\epsilon$ divergences cancel against one-loop insertions of the $d_i$ and $l_i$ LECs fixes $\\alpha_4$ and provides internal checks of the calculation.","core_discovery":"The central claim is that at order $q^5$ the full two-loop contribution to $g_A$ is captured by the coefficients $\\alpha_4 = -\\frac{7}{3}g_0(1-g_0^2)+16g_0^5$, with $\\gamma_4$ and $\\beta_4$ as given in Eqs. (32) and (34). The coefficient $\\alpha_4$ is universal (the same in EOMS and IR up to analytic terms), its $g_0^5$ term agrees with the earlier heavy-baryon renormalization-group result after correcting a factor of two in $\\tilde\\alpha_4 = 2\\alpha_2$, and the $\\log^2 M$ and $\\log M$ divergences cancel against one-loop insertions of the dimension-three low-energy constants. The regular pieces of the two-loop integrals are absorbed into the local counterterms $C$, so the difference between the two schemes appears only in $\\beta_4$. With the two LEC sets from pion-nucleon scattering the resulting correction is moderate, $2.6\\%$ for set 1 and $-17.4\\%$ for set 2 at the physical pion mass, which the authors take as evidence that the chiral expansion of $g_A$ converges reasonably well at fourth order, though they stress that the order-$q^6$ two-loop diagrams and $M^5$ corrections remain to be worked out.","pith_inferences":["A natural extension would be to fit the new closed forms directly to $g_A$ lattice data as a function of pion mass, which could independently determine the combination of $\\bar d_i$ LECs that enters $\\gamma_4$ and $\\beta_4$ and cross-check the pion-nucleon scattering values.","If the in-progress $q^6$ and $M^5$ corrections turn out to be as small as the $M^4$ term for LEC set 1, this would strengthen the case that the chiral expansion of $g_A$ is under control near the physical point; if they are large for set 2, the moderate-$M^4$ conclusion would need to be revised.","Because only $\\beta_4$ differs between the EOMS and IR schemes (the difference being absorbed into the counterterm $C_0$), the scheme dependence of the full two-loop prediction could be used as a diagnostic for missing higher-order terms when comparing with precise lattice data."],"forward_implications":["The scheme-independent coefficient $\\alpha_4$ fixes the leading non-analytic two-loop correction to $g_A$ in any chiral extrapolation, so future lattice-QCD extrapolations of the axial coupling can be carried out with two-loop covariant control through order $M^4$.","With LEC set 1 the fourth-order correction is only $2.6\\%$ at the physical pion mass, supporting the use of covariant baryon chiral perturbation theory for $g_A$ up to pion masses of about 300 MeV.","With LEC set 2 the same correction is $-17.4\\%$, so the claimed 'moderate' size depends on which pion-nucleon scattering LEC set is used; the expansion is not uniformly small.","The order-$q^6$ two-loop diagrams with $L_{\\pi N}^{(2)}$ vertices and the $M^5$ corrections are still needed before final conclusions; the paper explicitly leaves both to future work."],"supporting_citations":[{"why":"earlier heavy-baryon two-loop result; supplies the $g_0^5$ term and the $\\tilde\\alpha_4=2\\alpha_2$ comparison that this paper corrects.","marker":"[14]"},{"why":"first one-loop determination of $\\alpha_2$ and $\\beta_2$, the starting point for the chiral expansion.","marker":"[15]"},{"why":"completed the one-loop order-$q^4$ calculation of $g_A$, giving the $\\alpha_3$ coefficient used here.","marker":"[16]"},{"why":"provides the operator basis and the divergent parts $\\delta_i$ of the dimension-three low-energy constants used in the renormalization.","marker":"[19]"},{"why":"introduces the infrared/mixed/regular split of two-loop integrals and gives the two-loop nucleon self-energy used for $Z_{2-\\text{loop}}$.","marker":"[21]"},{"why":"gives the all-order-in-$M$ expressions for the sunset-type master integrals used to check the $\\epsilon$ and $M$ expansions.","marker":"[30]"},{"why":"combined analysis of elastic and inelastic pion-nucleon scattering; supplies the two sets of $c_i$ and $d_i$ LECs used in the numerical evaluation.","marker":"[36]"},{"why":"recent lattice determination in which $g_0$ comes out between 1.26 and 1.30, the comparison point for the chiral-limit value.","marker":"[9]"}],"fun_headline_variants":["Two-loop gA shifts: 2.6% to -17.4%","Covariant two-loop gA: first complete M^4 coefficients","gA two-loop corrections: moderate, scheme-dependent","Two-loop gA: corrections within -17.4% to 2.6%","New two-loop gA: universal coefficient, moderate shifts"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The numerical conclusion that the two-loop corrections are moderate assumes that the omitted order-$q^6$ two-loop diagrams with $L_{\\pi N}^{(2)}$ vertices and the $M^5$ corrections are small; the paper explicitly states that these pieces are still in progress and must be worked out before final conclusions.","fun_headline_variants_meta":{"raw":{"variants":["Two-loop gA shifts: 2.6% to -17.4%","Covariant two-loop gA: first complete M^4 coefficients","gA two-loop corrections: moderate, scheme-dependent","Two-loop gA: corrections within -17.4% to 2.6%","New two-loop gA: universal coefficient, moderate shifts"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000481,"raw_usage":{"total_tokens":2351,"prompt_tokens":888,"completion_tokens":1463,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":504,"completion_tokens_details":{"reasoning_tokens":1367}},"tokens_in":504,"tokens_out":1463,"duration_ms":11445,"temperature":1.0,"reasoning_tokens":1367,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-15T22:53:00.401917+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Compute the order-$q^6$ two-loop diagrams with $L_{\\pi N}^{(2)}$ vertices and the $M^5$ corrections to $g_A$: if their combined contribution at the physical pion mass is comparable to or larger than the reported $|\\Delta^{(4)}|$ (in particular for LEC set 2, where $\\Delta^{(4)}=-17.4\\%$), the claim that the leading two-loop corrections are moderate is falsified. A complementary check is a high-precision lattice value of $g_A$ at the physical pion mass that cannot be accommodated once these higher-order terms are included at their natural size.","supporting_citations":[{"cited_title":"The nucleon axial-vector coupling beyond one loop","cited_arxiv_id":"hep-lat/0605010","evidence_quote":"earlier heavy-baryon two-loop result; supplies the $g_0^5$ term and the $\\tilde\\alpha_4=2\\alpha_2$ comparison that this paper corrects."},{"cited_title":"Bernard, N","cited_arxiv_id":null,"evidence_quote":"first one-loop determination of $\\alpha_2$ and $\\beta_2$, the starting point for the chiral expansion."},{"cited_title":"Field Redefinitions and Wave Function Renormalization to $O(p^4)$ in Heavy Baryon Chiral Perturbation Theory","cited_arxiv_id":"hep-ph/9901235","evidence_quote":"completed the one-loop order-$q^4$ calculation of $g_A$, giving the $\\alpha_3$ coefficient used here."},{"cited_title":"Infrared renormalization of two-loop integrals and the chiral expansion of the nucleon mass","cited_arxiv_id":"0707.4296","evidence_quote":"introduces the infrared/mixed/regular split of two-loop integrals and gives the two-loop nucleon self-energy used for $Z_{2-\\text{loop}}$."},{"cited_title":"Ananthanarayan, S","cited_arxiv_id":null,"evidence_quote":"gives the all-order-in-$M$ expressions for the sunset-type master integrals used to check the $\\epsilon$ and $M$ expansions."}],"review_version":1}