{"id":"4cd87f32-04cf-458b-9d2d-f71c26a649af","arxiv_id":"2411.10106","paper_version":2,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":0,"one_line_summary":"Four-loop MSbar expressions are computed for all single-scale quark two-point Green's functions with a second-moment twist-2 Wilson operator insertion, including operator mixing and operator correlation functions.","lead":"This paper computes four-loop quantum corrections to quark matrix elements of the twist-2 Wilson operator used in parton distribution studies. The high-order results give lattice QCD a more precise perturbative benchmark for matching hadron structure measurements.","discovery_kind":"extension","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The new four-loop finite parts are not independently verified; every quoted check constrains poles or lower orders, so a Forcer/Feynman-rule bug would corrupt the central result while passing all checks.","rationale":"The paper is a genuine large-scale computation: more than 30,000 four-loop diagrams, public Forcer, reproduction of the known four-loop gamma_11, and explicit verification of operator mixing relations. The three-loop correlation functions are reproduced from [34], and the correlators are checked to be gauge independent. The central risk is not an obvious internal inconsistency - I found none - but the fact that the genuinely new O(a^4) finite parts are validated only by the integrity of the unshown pipeline. The checks are necessary but not sufficient: the pole parts of the amplitudes determine the anomalous dimensions, and the lower-order terms are fixed by previous work, so a bug affecting only the O(epsilon^0) coefficient of a four-loop integral would evade every check reported. Since the manuscript does not ship the Forcer input scripts or the diagram-generation setup, an independent group cannot reproduce the result without reconstructing the entire pipeline. This justifies the reader's CONDITIONAL verdict: the result is plausible and well cross-checked internally, but the new finite parts should be independently reproduced (or the pipeline released) before the 'first four-loop expression' claim is fully accepted. My read does not change the reader's verdict, hence UNCHANGED.","tokens_in":27615,"tokens_out":12924,"duration_ms":130940,"concrete_test":"Independently recompute one new four-loop non-forward form factor, e.g. bSigma^(1)_1 in the Landau gauge for SU(3), from scratch: generate the 12878 four-loop graphs of Table 1 with Qgraf using the operator Feynman rules of Section 2, reduce them with a fresh Forcer run, and compare the O(a^4) finite term at p^2 = mu^2 with Eq. (3.11) (-100557.006490 for bSigma^(1)_1). A matching result would settle the concern; a mismatch would invalidate the central claim. As a cheaper preliminary, also truncate the ancillary four-loop expressions to O(a^3) and check them against the published three-loop asymmetric results of refs. [26,27,28].","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central claim is the four-loop finite parts of the operator matrix elements and correlation functions. The paper's verification consists of: reproducing the four-loop anomalous dimension gamma_11 from pole parts; checking gamma_12 = -1/2 gamma_11 and gamma_22 = 0; reproducing the three-loop correlation functions of [34]; and confirming alpha-independence of the correlators. None of these tests constrains the O(epsilon^0) coefficients of the newly evaluated four-loop master integrals, because the pole structure is insensitive to the finite part of those integrals and the three-loop checks do not exercise the genuinely new a^4 terms. For the asymmetric (b and e) configurations, which are the non-forward generalizations of [23], no explicit three-loop comparison with refs. [26,27,28] is reported, even though those references cover the same objects at three loops. The Forcer package is public, but the manuscript does not ship the diagram-generation and reduction scripts; only the output is provided. A bug in the user's implementation of the non-zero operator-momentum Feynman rules, or in Forcer's reduction of the relevant integrals, would therefore produce wrong four-loop finite parts while leaving every reported check unchanged. This is the load-bearing assumption behind the 'first four-loop expression' claim.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"This paper presents a four-loop computation, in the MS scheme and the massless limit, of the Green's functions obtained by inserting the second moment of the twist-2 flavour nonsinglet Wilson operator into a quark two-point function, in all three single-scale momentum routings: the symmetric insertion G, and two asymmetric routings bG and eG in which momentum flows out through the operator itself. Operator mixing with the total-derivative operator O_2^{μν} is included for the asymmetric configurations. The paper also computes the correlation functions Π_ij of the two gauge-invariant operators to four loops, including contact-term renormalization. Results are presented as analytic expressions for general colour group and linear-covariant-gauge parameter, with Landau-gauge form factors in Appendix A, and as numerical SU(3) expressions. The author reproduces the four-loop anomalous dimension γ_11, verifies γ_12 = −γ_11/2 and γ_22 = 0 at four loops, reproduces the three-loop correlation functions of [34], and checks the gauge independence of the operator correlation functions. The new four-loop O(ε^0) finite parts are claimed as the first results of this order for these objects.","tokens_in":27835,"tokens_out":12771,"duration_ms":130410,"significance":"The computation is a direct Feynman-graph evaluation with no free parameters and no reverse-engineered constants: every quoted check anchors the poles or lower orders externally, and the manuscript is explicit about what has been verified. If correct, the four-loop finite parts are of genuine reference value for precision matching between continuum MS-scheme results and lattice determinations, and the decomposition into the Lorentz bases of (3.2) and (4.3) is appropriate for RI'-scheme conversions. Strengths include the reproduction of γ_11 against [5,6], the non-trivial verification of the mixing relations γ_12 = −γ_11/2 and γ_22 = 0, the three-loop comparison for the correlation functions, and the α-independence check. The central weakness, however, is that the genuinely new a^4 finite parts are not independently checked: the reported tests constrain pole terms and lower orders only, so a systematic error in the new master integrals or in the operator Feynman rules would pass every check listed in Sections 3 and 4. Because the asymmetric configurations lack the three-loop comparison that the correlators receive, this verification gap lands precisely on the paper's headline claim.","major_comments":[{"comment":"The verification reported for the new four-loop results constrains only the pole structure and lower orders, not the new O(ε^0) finite parts. Reproducing γ_11 from the 1/ε poles, verifying γ_12 = −γ_11/2 and γ_22 = 0, and reproducing the three-loop correlators of [34] do not exercise the a^4 finite terms that constitute the paper's central claim; the gauge-independence check is likewise insensitive to colour- and α-independent errors in those terms. I ask for an additional check that directly constrains the new finite parts. Concretely, one of the following would suffice: (i) an independent evaluation of a subset of the new four-loop master integrals by a second method (e.g., an alternative reduction or high-precision numerical integration); (ii) a second, independent code path for at least one of the new form factors at a^4; or (iii) a detailed three-loop comparison of the b/e form factors with refs. [26-28] together with the release of the reduction inputs. As it stands, the central claim rests on the correctness of a single reduction pipeline.","section":"Section 3, Eqs. (3.4)-(3.7); Section 4, Eqs. (4.8)-(4.15)"},{"comment":"No comparison with the existing three-loop non-forward results [26-28] is reported for the asymmetric (b and e) configurations, even though the introduction cites these references as the three-loop state of the art for non-forward Wilson-operator matrix elements. The b/e form factors are precisely the non-forward generalizations of [23] in which the operator Feynman rules at non-zero operator momentum and the asymmetric routing enter; a three-loop comparison would therefore test exactly the ingredients that are new at four loops. The paper explicitly reports the analogous check for the correlation functions ('We have reproduced the previous three loop expressions, [34]'), and the eΣ and bΣ expressions in the Landau gauge should similarly be checked against [26-28] where the momentum configurations overlap. This is load-bearing because the b/e form factors appear only as bare output in the main text and Appendix A.","section":"Section 3, Eqs. (3.10), (3.11); Appendix A, Eqs. (A.1)-(A.3)"},{"comment":"The ancillary file contains the final expressions but not the computational workflow: the manuscript does not state the versions of Qgraf, Form, and Forcer used, nor does it provide the diagram-generation scripts, the operator insertion rules for non-zero operator momentum, or the reduction inputs. A bug in the user-supplied operator Feynman rules or in the reduction setup would corrupt exactly the new finite parts while leaving every reported check unchanged, since those checks are insensitive to the finite parts of the new four-loop integrals. Please either include the run scripts and input files in the ancillary material, or specify the exact software versions and enough configuration detail to allow independent reproduction of the new a^4 coefficients.","section":"Data Availability Statement; Section 3"}],"minor_comments":[{"comment":"The sentence 'γij(a) are given by (3.4), (3.6) and 3.7)' has a missing parenthesis before the last reference; it should read '(3.7)'.","section":"Section 4, text before Eq. (4.11)"},{"comment":"The SU(3) a^4 coefficient of bΣ_2^(2) is printed as 2014630.937750, roughly an order of magnitude larger than the neighbouring a^4 coefficients (for comparison, bΣ_1^(2) gives 145064.413688); please confirm that this value, and the varying decimal precision of the printed coefficients (six to eight digits), coincide with the analytic expressions in the ancillary file.","section":"Eq. (3.11)"},{"comment":"The abstract and the body use both 'MSbar' and 'MS' for the renormalization scheme; since the finite parts of operator Green's functions are scheme-dependent, a single notation should be fixed (the body uses MS throughout).","section":"Abstract and Section 1"}],"recommendation":"major_revision","confidential_remarks":"The paper is a single-author computation extending the author's own three-loop programme, and it is honest about what has been checked. The main risk is not internal inconsistency but that the new four-loop finite parts — the headline result — are verified neither by the reported checks nor by an independent second computation. The natural remedy is to require, in revision, the three-loop comparison of the b/e form factors against refs. [26-28] and the release of the diagram-generation/reduction scripts, or an independent check of a subset of the new master integrals. I would not recommend rejection: the computation is plausible, the checks that exist are real, and the requested additions are within the manuscript's scope. The paper fits the journal's scope as a high-order perturbative QCD computation of reference value."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"This is a careful, substantial computation: first four-loop results for the second-moment twist-2 nonsinglet Wilson operator inserted in a quark 2-point function in all three single-scale momentum configurations, plus four-loop correlators of both operators. The paper reproduces the known four-loop gamma_11, verifies the gamma_12 = -gamma_11/2 and gamma_22 = 0 relations, and gets the three-loop correlators of [34] back. The gauge-independence check on the correlators is also good. This is real work and the analytic results are in an ancillary file, which is more than many papers in this area provide.\n\nThe soft spot the stress-test flag is real. The checks all constrain either pole parts or lower-loop results. Reproducing gamma_11 from the symmetric insertion confirms the Forcer pipeline for those integrals at the pole level, but not the O(epsilon^0) coefficients that are the actual new output. Reprocessing the three-loop correlators exercises the same machinery at lower order. None of this would catch a bug in the new four-loop finite parts, especially in the asymmetric b and e configurations where a non-zero operator momentum forces a different set of Feynman rules. The paper does not report any explicit three-loop comparison with refs. [26,27,28], even though those papers cover the same asymmetric objects at three loops. That omission is the biggest practical concern. It would be easy to fix: just display the three-loop terms from the appendix against the published expressions. The code is not shipped either, so the reader cannot trace the finite parts independently.\n\nThat said, this is not a fatal flaw. The internal consistency is decent, the renormalization-group structure is properly handled, and the paper is honest about what it does and does not claim. The lack of an independent four-loop finite check is typical for Forcer-based results at this frontier; you either trust the package or you don't, and the author has a long track record with these tools. I would send this to a referee knowledgeable in multiloop computations. The referee should ask for the missing three-loop comparisons and, ideally, one additional finite-order cross-check (a low-N_f limit, a small color group, or a second momentum configuration that can be derived from the first). If the author supplies those, the paper is publishable as is.\n\nWho is this for? Lattice QCD practitioners doing RI'-to-MS matching for PDF moments, and perturbative hadron-structure people who need four-loop Wilson-operator matrix elements. It is incremental, not paradigm-shifting, but it is exactly the kind of precise input that matching calculations need.","headline":"A solid four-loop extension of the n=2 Wilson operator matrix elements, but the new finite parts rest on an unshipped Forcer pipeline and the reported checks do not directly constrain them.","tokens_in":28299,"tokens_out":2917,"would_cite":true,"duration_ms":32144,"reading_group":"maybe","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":[],"pacs":["12.38.Bx","11.10.Gh","12.38.-t"],"model":"deepseek-v4-flash","headline":"This paper claims the first four-loop Green's functions for the second-moment twist-2 Wilson operator in a quark two-point function, in all three single-scale momentum configurations, plus four-loop correlation functions of the operator…","keywords":["four-loop QCD","twist-2 Wilson operator","operator matrix elements","renormalization","lattice QCD matching","massless Feynman integrals","MSbar scheme","anomalous dimensions"],"falsifier":"Recompute one of the new four-loop finite parts, say the SU(3) form factor $\\Sigma^{(1)}_1$ at $N_f = 3$, with an independent integral-reduction algorithm and require the resulting anomalous dimension to match $\\gamma_{11}$; any disagreement in the finite part would falsify the paper's four-loop claim.","tokens_in":27416,"feed_emoji":"🧮","tokens_out":7046,"duration_ms":68020,"temperature":0.7,"pith_summary":"The paper's aim is to obtain the first four-loop expressions, in the MSbar scheme and the massless limit, for the Green's functions obtained by inserting the second moment of the twist-2 flavour non-singlet Wilson operator into a quark two-point function, covering all three single-scale momentum configurations: zero momentum at the insertion, and the two asymmetric routings where a non-zero momentum flows through the operator and one quark leg is nullified. It also computes the four-loop correlation functions of the two gauge-invariant operators in the same scheme. These matrix elements are the continuum objects that lattice QCD determinations of parton distributions must match onto at high energy, so one more perturbative order reduces the systematic uncertainty in that matching. The author deliberately keeps a non-zero gauge parameter as a consistency check and provides the full Lorentz decomposition, which is enough to construct any lattice RI' matching scheme.","feed_headline":"First four-loop twist-2 operator matrix elements computed","feed_subtitle":"New perturbative results give lattice QCD the matching data to sharpen parton distribution extractions.","key_machinery":"The argument is carried by the automatic evaluation of massless four-loop two-point Feynman integrals in dimensional regularization, implemented in the Forcer package and driven by Qgraf for graph generation. The operator insertions are projected onto a complete basis of two symmetric traceless tensors, $T_1$ and $T_2$, so each Green's function reduces to scalar form factors; the asymmetric configurations require the upper-triangular mixing matrix between $O_1$ and $O_2$, including the total derivative operator that is physical with zero anomalous dimension. The correlation functions are decomposed into a basis of three Lorentz tensors and require additional contact renormalization constants, whose renormalization group functions are extracted from the four-loop expressions.","core_discovery":"The central claim is that the complete set of scalar form factors for the operator pair $O_1 = S(\\bar\\psi\\gamma^\\mu D^\\nu\\psi)$ and $O_2 = S\\partial^\\mu(\\bar\\psi\\gamma^\\nu\\psi)$ is determined to four loops in the MSbar scheme: the zero-momentum insertion reproduces the known four-loop anomalous dimension, the two asymmetric insertions determine the off-diagonal mixing renormalization constant, and the operator correlation functions fix the contact renormalization functions to the same order. These are the first four-loop results for these Green's functions, extending the three-loop forward matrix elements and three-loop correlation functions by one order. The results are analytic in the linear covariant gauge parameter and in the colour factors, with SU(3) numerical values provided.","pith_inferences":["If confirmed, the same approach can be aimed at transversity and other spin-dependent twist-2 operators, which the paper names as the natural next target.","The four-loop relation $\\gamma_{12} = -\\tfrac{1}{2}\\gamma_{11}$ supplies a strong consistency condition that any future five-loop computation of these operators should satisfy.","A lattice collaboration could combine the four-loop correlation functions with the forward matrix elements to cross-check RI' to MSbar conversion factors without re-deriving the operator renormalization.","Because the gauge parameter is kept non-zero throughout, one could test gauge invariance of physical combinations at four loops, a check that was previously only available at three loops."],"forward_implications":["The zero-momentum form factors provide an independent four-loop confirmation of the non-singlet operator anomalous dimension.","The asymmetric form factors give the four-loop off-diagonal mixing constant $Z_{12}$ and verify $\\gamma_{12} = -\\tfrac{1}{2}\\gamma_{11}$.","The gauge-invariant operator correlation functions supply four-loop matching data that do not require a gauge-fixed lattice action.","Setting $\\alpha = 0$ in the analytic expressions yields Landau-gauge matrix elements for direct lattice comparison.","The full four-loop form factors allow lattice collaborations to define any RI' projection scheme and convert the result to MSbar."],"supporting_citations":[{"why":"Supplies the Forcer algorithm that evaluates all massless four-loop two-point integrals used in the computation.","marker":"[12, 13]"},{"why":"Gives the three-loop forward matrix element that this paper extends to four loops.","marker":"[23]"},{"why":"Gives the three-loop operator correlation functions and contact-term formalism extended here.","marker":"[34]"},{"why":"Provide the four-loop low-moment operator anomalous dimensions reproduced as a check.","marker":"[5, 6]"},{"why":"Provide the three-loop and off-shell anomalous dimensions used to validate the renormalization framework.","marker":"[1, 2, 3, 4]"},{"why":"Defines the MS scheme renormalization used for the gauge-independent operators.","marker":"[35]"},{"why":"Qgraf generates the Feynman graphs enumerated in Tables 1 and 2.","marker":"[36]"},{"why":"Introduce the RI' scheme that motivates the lattice matching context.","marker":"[29, 30]"}],"fun_headline_variants":["First four-loop twist-2 operator matrix elements","Complete four-loop twist-2 operator form factors","Four-loop twist-2 operators: all form factors computed","New four-loop results for moment-2 Wilson operators"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"Everything new rests on the assumption that the automatic four-loop integral engine evaluates correctly all of the more than 32,000 Feynman graphs listed in the tables, since only the anomalous dimensions and the three-loop limits are independently verified.","fun_headline_variants_meta":{"raw":{"variants":["First four-loop twist-2 operator matrix elements","Complete four-loop twist-2 operator form factors","Four-loop twist-2 operators: all form factors computed","New four-loop results for moment-2 Wilson operators"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.001123,"raw_usage":{"total_tokens":4606,"prompt_tokens":816,"completion_tokens":3790,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":432,"completion_tokens_details":{"reasoning_tokens":3728}},"tokens_in":432,"tokens_out":3790,"duration_ms":28669,"temperature":1.0,"reasoning_tokens":3728,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-12T19:57:26.363851+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Recompute one of the new four-loop finite parts, say the SU(3) form factor $\\Sigma^{(1)}_1$ at $N_f = 3$, with an independent integral-reduction algorithm and require the resulting anomalous dimension to match $\\gamma_{11}$; any disagreement in the finite part would falsify the paper's four-loop claim.","supporting_citations":[{"cited_title":"G¨ ockeler, R","cited_arxiv_id":null,"evidence_quote":"Gives the three-loop forward matrix element that this paper extends to four loops."},{"cited_title":"Gracey, Eur","cited_arxiv_id":null,"evidence_quote":"Gives the three-loop operator correlation functions and contact-term formalism extended here."},{"cited_title":"Gracey, J","cited_arxiv_id":null,"evidence_quote":"Defines the MS scheme renormalization used for the gauge-independent operators."}],"review_version":1}