{"id":"c02a84a6-e2c6-4284-8d3b-4d0da320719e","arxiv_id":"2506.21013","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":3,"one_line_summary":"Derives the NLO (1/Nc) chiral-odd GPDs in the pion mean-field picture, proves polynomiality and sum rules, and gives gradient-expansion estimates partially matching lattice QCD.","lead":"This paper computes the next-to-leading-order corrections to the nucleon's chiral-odd generalized parton distributions in the large-Nc limit of QCD, completing their spin-flavor structure. The results provide nonperturbative predictions needed for planned experiments at Jefferson Lab, COMPASS, and the Electron-Ion Collider.","discovery_kind":"extension","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The Fig. 6 lattice agreement rests on an uncontrolled truncation: with the adopted profile (113) and MR₀=1, the gradient expansion parameter in (106) is O(1) (|∇U|/M≈2), so omitted UV-finite and discrete-level contributions are not parametrically suppressed; the agreement claim is therefore not…","rationale":"The paper extends the large-N_c pion mean-field framework for chiral-odd GPDs to NLO in 1/N_c, deriving spin-flavor structures, multipole mean-field GPDs, polynomiality, and sum rules, plus gradient-expansion numerics with a lattice comparison. For the central claim to hold, (a) the formal derivation must be internally consistent and match the general constraints of Sec. II, and (b) the retained Dirac-sea estimate must dominate the omitted contributions, or the comparison must be robust to their inclusion. I checked (a) first because it is the structural core. The moment formula (68), the G5 and parity restrictions (85)–(88), the grand-spin selection rules (80)–(83), and the bounds (90)–(92), (B9), (B16), (B27) produce polynomials in ξ whose maximal degrees (m+1 for the monopole via the ξ²E_T piece, m−1 or m for the dipoles, m−1 for the quadrupole) match what Eq. (4) requires for the combinations (64). The sum rules (100), (C6), (C11), (C17) agree with the first-moment relations (6) and the form-factor combinations (A3). The N_c scalings (61)–(62) are mutually consistent provided the nontrivial relation (63) cancels the would-be N_c⁴ terms in (64b), which is cited to Ref. [1]. I found no internal inconsistency in the formal claim. The insecure part is (b). The paper is transparent: Sec. VIII states that only UV-divergent gradient-expansion contributions were computed and that a full single-particle sum is needed, and footnote 2 flags the matching subtlety. But the abstract nonetheless asserts 'good agreement with lattice QCD predictions.' My concern is that the truncation is not merely incomplete but uncontrolled: the gradient expansion parameter is O(1) for the adopted profile (MR₀=1), there is no large logarithm justifying retention of only divergent pieces, the discrete level is known to contribute sizeably to other observables in this model, the identification of Z_{mf,1} with ĒT̄^{u−d}+O(UV-finite) rests on the unpublished Ref. [81], and ξ=1/3 lies outside the formal O(N_c^{−1}) kinematics. A secondary issue is scale: the model outputs are at μ≈0.6 GeV while the lattice results of Ref. [52] are at ≈2 GeV, and no evolution is discussed. These are all addressable — the full single-particle sum is a standard computation in this framework — but none is addressed here. This matches the reader's weakest assumption; my contribution is the concrete observation that the needed parametric suppression is in fact absent, which strengthens rather than weakens the conditionality. The formal part, by contrast, has independent support from discrete symmetries and passes consistency checks against the general constraints. The verdict should remain conditional: the framework and structural claims are acceptable, but the numerical and lattice claims need either the full computation or explicit requalification.","tokens_in":34690,"tokens_out":21390,"duration_ms":229703,"concrete_test":"Evaluate the NLO contribution to ĒT̄^{u−d} with the full single-particle sum (discrete level plus Dirac sea) in Eq. (48)/(65), using the same self-consistent profile and parameters (M=350 MeV, MR₀=1, form factor (115)), at ξ=1/3, t=−1.02 GeV². If the full result deviates from the truncated UV-divergent curve of Fig. 6 by more than the plotted agreement (e.g., peak magnitude shifts by >30% or the negative-x shape changes qualitatively), the truncation is uncontrolled and the 'good agreement' claim must be withdrawn or requalified. A cheaper diagnostic: compute the α=2 two-chiral-field term in the gradient expansion; if it is comparable to the α=0+1 result at these kinematics, the expansion is already uncontrolled within the gradient approximation.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central numerical claim — 'good agreement with lattice QCD' (abstract; Sec. VII, Fig. 6) for Ă̄ĒT̄^{u−d} at ξ=1/3, t=−1.02 GeV² — requires that the retained UV-divergent α=0,1 gradient-expansion terms dominate the omitted UV-finite multi-chiral-field terms (Sec. VI B) and the discrete-level/connected-diagram contribution (Secs. VII–VIII). Three specific points undermine this. (i) The expansion (106) in powers of iMγ·∇Uγ⁵, stated to require ∂U ≪ M, is not controlled for the adopted profile (113): P(r)=−2arctan(R₀²/r²) with MR₀=1 gives |∇U|≈2/R₀=2M at r≈R₀. With Λ=2^{1/2}ρ̄⁻¹≈850 MeV (115) and M=350 MeV (114), the logarithmic enhancement log(Λ/M)≈0.9 is not large; α≥2 and UV-finite terms are O(1) relative to the retained pieces, so the truncation is a model choice, not a systematic expansion. (ii) The interpretation Z_{mf,1}=ĒT̄^{u−d}+O(UV-finite) (footnote 2) requires H_T^{u−d} to have no UV-divergent contribution at this order; this is attributed to the unpublished Ref. [81]. If H_T^{u−d} carried a UV-divergent piece, Fig. 6 would compare the wrong object. (iii) The derivation assumes ξ=O(N_c^{−1}) and x=O(N_c^{−1}) (Sec. IV C), yet Fig. 6 is evaluated at ξ=1/3, far outside that regime, and Sec. VIII concedes that a full single-particle sum is required for complete predictions. The formal results (spin-flavor structures, polynomiality, sum rules) are internally consistent and defensible, but the advertised numerical comparison is not established by the present truncation.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"This paper extends the large-Nc pion mean-field (chiral quark-soliton) description of nucleon chiral-odd GPDs from leading order to next-to-leading order in the 1/Nc expansion. The NLO baryon matrix element of the nonlocal chiral-odd operator is derived in the single-particle representation, including the first-order rotational zero-mode correction; its spin-flavor structure is decomposed into multipole mean-field GPDs (monopole Zmf,0, dipoles Zmf,1 and ~Zmf,1, quadrupole Zmf,2) through a two-dimensional multipole expansion in the transverse momentum transfer, and linear relations (64) between these objects and the standard GPDs HT, ~HT, ET, and ~ET are established. The m-th moments of the mean-field GPDs are shown to be polynomials in ξ with the correct power structure, using G5 symmetry, parity, and grand-spin selection rules, and the first moments are matched to the nucleon tensor form factors through sum rules. Numerical estimates are then obtained by gradient-expanding the quark propagator and retaining only the UV-divergent α = 0 and α = 1 contributions; the resulting ~ET^{u-d} is compared with lattice QCD data in Fig. 6, with the abstract reporting good agreement. The paper states in Secs. VI and VIII that a complete prediction requires the full single-particle sum, which is not performed.","tokens_in":35134,"tokens_out":14151,"duration_ms":128859,"significance":"If the formal derivations are correct, this is a useful and substantial extension of Ref. [1]: it completes the spin-flavor classification of chiral-odd GPDs in the mean-field picture, provides explicit polynomiality proofs at the level of the single-particle representation, and yields testable large-Nc predictions including the hierarchy (62) and the relation (63). The proofs are grounded in the discrete symmetries of the mean-field Hamiltonian rather than in any fit to the GPD data, and the parameters (M, R0, Λ) are fixed by the instanton-vacuum model before the lattice comparison, so the formal section is free of circularity. The numerical section, by contrast, is not a controlled calculation: as the paper itself concedes, the retained UV-divergent gradient-expansion terms do not constitute the full result, and the truncation parameter is not small at the adopted parameters. The lattice comparison in Fig. 6 should therefore be read as an illustration of the mechanism rather than as an established quantitative prediction.","major_comments":[{"comment":"The truncation of the gradient expansion is uncontrolled at the adopted parameters, so the 'good agreement with lattice QCD' claim in the abstract and in Sec. VII is not established. The expansion (106) is in powers of iMγ·∇Uγ5 and is stated to require ∂U ≪ M, but with the profile (113) and MR0 = 1 one has |∇U| ≈ |P'(r)| ≈ 2/R0 = 2M at r ≈ R0, i.e., the expansion parameter is O(1), not small. With Λ ≈ 835 MeV and M = 350 MeV from Eqs. (114)-(115), the logarithmic enhancement log(Λ/M) ≈ 0.9 is too small to compensate, so the UV-finite terms explicitly dropped in Eqs. (111)-(112) and the α ≥ 2 terms are of the same order as the retained UV-divergent pieces. The paper itself states in Sec. VI that a complete prediction requires summing over all single-particle wave functions and in Sec. VIII that the study was limited to UV-divergent contributions and that the discrete-level (connected-diagram) contribution is missing; given an O(1) expansion parameter, these omissions cannot be assumed small, and the comparison in Fig. 6 is made without any estimate of the systematic error.","section":"Sec. VI, Eqs. (106), (113)-(115), and Fig. 6"},{"comment":"The numerical comparison is made outside the kinematic regime in which the mean-field GPD basis was derived. The relations (64) and the identification of the computed objects with standard GPDs assume x = O(Nc^{-1}) and ξ = O(Nc^{-1}) from Eqs. (53) and (55), but Fig. 6 is evaluated at ξ = 1/3, for which Nc ξ = O(1), and the plotted range extends to |x| = 1. Furthermore, Sec. VI states that the retained UV-divergent contributions live only in the ERBL region |x| < ξ; if that is so, the curve in Fig. 6 should vanish for |x| > 1/3, and the claimed agreement in the negative-x region needs to be restricted to that interval. Please clarify the x-support of the plotted curves and either restrict the comparison to the parametric regime or justify the continued use of Eq. (64) at ξ = O(1).","section":"Sec. IV C, Sec. VII, Eqs. (53)-(55), Fig. 6"},{"comment":"The identification of the computed object with ~ET^{u-d} rests on an unpublished reference. Footnote 2 states that Zmf,1 = ~ET^{u-d} + O(UV-finite) because H_T^{u-d} has no UV-divergent contribution at this order, citing the 'in preparation' Ref. [81]. This assertion is load-bearing: if H_T^{u-d} did carry a UV-divergent piece, Fig. 6 would compare the wrong combination of GPDs. Since Ref. [81] is not available, the authors should either provide the argument in this paper, replace the citation with an available source, or present the lattice comparison in a form that does not depend on this unverified step.","section":"Footnote 2, Sec. VII, Ref. [81]"}],"minor_comments":[{"comment":"There are numerous typos, e.g., 'adotped' in Sec. VII, 'shoud' in footnote 2, 'prepration' in Ref. [81], and 'emergies' in Sec. IV D; the manuscript would benefit from a careful proofreading pass.","section":"Throughout"},{"comment":"Eq. (106) presents the propagator expansion without the symmetrization introduced in the footnote; please make the presentation consistent so that the reader can see that the truncation is applied to the symmetrized operator.","section":"Sec. VI, Eqs. (105)-(106)"},{"comment":"The shorthand '1' for the unit spin-flavor structure in Eq. (59) is confusing when it appears in expressions such as Eq. (46); consider using an explicit delta symbol instead.","section":"Sec. IV D, Eq. (59)"},{"comment":"The statement that one 'observes a vanishing ~ET GPD' should be qualified directly in the main text: the vanishing follows from the relation (118), which holds only at the level of UV-divergent accuracy, as the following sentences concede.","section":"Sec. VII, Eq. (118)"},{"comment":"The sentence about higher-twist operators and gluon contributions ('gluon contributions ~ (M ρ-bar)^0 are not parametrically suppressed') is unclear in context, since the operators Γ = γ+, γ+γ5, iσ+j are all leading twist; please clarify which operators are being discussed.","section":"Sec. IV A"},{"comment":"Please indicate the statistical and systematic uncertainties of the lattice result in Fig. 6 and specify how the dot-dashed curve was obtained from the data points of Ref. [52].","section":"Sec. VII, Fig. 6"},{"comment":"The abstract states 'Our results show good agreement with lattice QCD predictions' without the qualifications given in Sec. VIII; please bring the abstract in line with the stated limitations of the gradient-expansion estimate.","section":"Abstract and Sec. VIII"}],"recommendation":"major_revision","confidential_remarks":"The formal half of the paper (Sections II-V and the appendices) is careful, internally consistent, and within the journal's scope; the NLO spin-flavor completion and the polynomiality/sum-rule proofs are publishable contributions even if the numerical section were substantially downgraded. The main risk is the reliance on the unpublished 'in preparation' Ref. [81] for a step on which the lattice comparison depends; I recommend that the editor require this to be resolved. The paper is honest about its limitations in Sec. VIII, but the abstract overstates the numerical result, so the revision should reconcile the two. I see no circularity or data-fitting issue in the formal claims."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Lead: the paper's formal core is solid and genuinely new: NLO spin-flavor structures for chiral-odd GPDs, the multipole mean-field basis in Eq. (64), and the polynomiality/sum-rule proofs via G5 and parity symmetries. That part deserves a serious referee. The numerical 'good agreement' with lattice, however, rests on a truncation that is not controlled, and the abstract overstates it.\n\nWhat is new: extending the LO analysis [1], Kim derives the complete NLO spin-flavor structures for flavor singlet and non-singlet, gives the Nc scalings (62), and proves polynomiality and first-moment sum rules at NLO. The multipole expansion in transverse momentum transfer is a useful organizing principle, and the Nc hierarchy H^{u+d}_T << bar_E_T^{u-d} << E^{u+d}_T emerges naturally. I could not verify every algebraic step, but the symmetry arguments are explicit and self-consistent.\n\nSoft spots: the gradient expansion (106) is truncated to the UV-divergent alpha=0,1 terms. With the adopted profile (113) P(r) = -2 arctan(R0^2/r^2) and MR0=1, the expansion parameter |grad U|/M is O(1) around r approx R0, so the omitted UV-finite terms and the discrete-level (connected) contribution are not parametrically suppressed. The paper acknowledges the omission in Sec. VIII, but the abstract and Fig. 6 still claim good agreement with lattice QCD. That agreement is a model result, not a systematic prediction. The comparison in Fig. 6 is at xi=1/3, which is far outside the xi=O(1/Nc) regime in which the kinematics were derived. A further load-bearing step is the identification of Z_mf,1 with bar_E_T^{u-d}, which assumes H_T^{u-d} has no UV-divergent piece; that fact is attributed to the unpublished Ref. [81]. The referee needs to see that calculation or see the claim softened.\n\nVerdict: the formal results are defensible and useful for the large-Nc community. The numerical curves are fine as model estimates if labeled as such. I would send this to peer review with a request to temper the abstract and to make the truncation assumptions explicit in the comparison. The paper is not a desk reject; it is a solid formal contribution with a numerical section that needs recalibration.","headline":"Solid NLO-large-Nc derivation of chiral-odd GPD structures; the advertised lattice agreement is not controlled and needs tempering.","tokens_in":35705,"tokens_out":4107,"would_cite":true,"duration_ms":43003,"reading_group":"maybe","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":[],"pacs":[],"model":"deepseek-v4-flash","headline":"The pion mean-field picture of large-$N_c$ QCD yields the complete next-to-leading-order chiral-odd GPDs, whose moments obey polynomiality and whose first moments reproduce the tensor form factor sum rules.","keywords":["chiral-odd GPDs","large-Nc QCD","pion mean-field","next-to-leading order","polynomiality","tensor form factors","gradient expansion","transversity"],"falsifier":"At $\\xi=1/3$ and $t=-1.02$ GeV$^2$, evaluate $\\bar E_T^{u-d}(x)$ using the complete sum over all single-particle states including the discrete level; if the positive-$x$ result does not move appreciably toward the lattice points while negative-$x$ agreement is preserved, the truncation to UV-divergent gradient terms is not the dominant approximation.","tokens_in":34454,"feed_emoji":"⚛️","tokens_out":8538,"duration_ms":87530,"temperature":0.7,"pith_summary":"The paper aims to establish that the pion mean-field picture of the nucleon in the large-$N_c$ limit of QCD can supply the complete next-to-leading-order ($1/N_c$) contributions to the nucleon's chiral-odd generalized parton distributions, filling the spin-flavor structures that were missing at leading order in both the flavor-singlet and flavor-non-singlet sectors. It derives the rotational zero-mode correction to the mean field, organizes the resulting matrix elements into multipole mean-field GPDs, and proves that their moments are polynomials in the skewness $\\xi$ with the standard degree bounds, with the first moments satisfying the tensor form factor sum rules. Using a gradient expansion that keeps the UV-divergent Dirac-sea terms, the paper produces numerical estimates for the non-forward region; the flavor-non-singlet combination $\\bar E_T^{u-d}$ agrees with available lattice QCD data in the negative-$x$ region. If this program is correct, the mean-field framework becomes a complete nonperturbative tool for transversely polarized quark distributions, with immediate use for exclusive meson production phenomenology at current and future facilities.","feed_headline":"Chiral-odd GPDs computed to next order in large-Nc QCD","feed_subtitle":"The new NLO terms obey polynomiality and sum rules, and match lattice QCD in the flavor-non-singlet sector.","key_machinery":"The load-bearing object is the pion mean-field picture of the nucleon in the large-$N_c$ limit: a static hedgehog pion field $U(x)=\\exp[i\\,\\hat{x}\\cdot\\tau\\,P(r)]$ with $N_c$ valence quarks and a distorted Dirac sea, quantized by collective flavor rotations with angular velocity $\\Omega_a = J_a/I$. The machinery that carries the argument is the NLO rotational correction to the matrix element of the effective chiral-odd operator, expressed in the single-particle representation with energy denominators from particle-hole excitations; the multipole expansion in the two-dimensional transverse momentum transfer then defines the multipole mean-field GPDs $Z_{mf,0}$, $Z_{mf,1}$, $\\tilde Z_{mf,1}$, $Z_{mf,2}$. Polynomiality is proven from three discrete ingredients: the $G_5$ transformation (time reversal combined with an isospin rotation), parity, and the grand-spin selection rules that truncate the partial-wave expansion of $e^{i\\Delta\\cdot \\hat{X}}$ to finite order. The gradient expansion of the quark propagator, keeping the single-chiral-field UV-divergent terms at order $\\alpha=0$ and $\\alpha=1$, provides the numerical estimates that are compared with lattice QCD.","core_discovery":"On the author's own terms, the central result is that at next-to-leading order in $1/N_c$ the rotational correction to the pion mean field generates exactly the spin-flavor multipole structures that leading order cannot: a dipole operator in the flavor-singlet channel and monopole plus quadrupole operators in the flavor-non-singlet channel. These enter through single-particle sums over occupied and non-occupied Dirac levels, with the moment of inertia $I$ setting the $N_c$ suppression, and they combine into four multipole mean-field GPDs $Z_{mf,0}$, $Z_{mf,1}$, $\\tilde Z_{mf,1}$, $Z_{mf,2}$ that map linearly onto the standard chiral-odd GPDs $H_T$, $\\tilde H_T$, $E_T$, $\\tilde E_T$. The paper proves that the $m$-th moments of these mean-field GPDs are even or odd polynomials in $\\xi$ of the required degree, using the $G_5$ (time-reversal times isospin) symmetry, parity, and grand-spin selection rules, and that the first moments reproduce the tensor form factors $H_T(t)$, $\\tilde H_T(t)$, $E_T(t)$, with $\\int dx\\, \\tilde E_T = 0$. Numerically, keeping the UV-divergent $\\alpha=0$ and $\\alpha=1$ gradient-expansion terms, all displayed GPDs are convex peak-like functions centered at $x=0$ with a smooth crossover at $x=\\pm\\xi$, the magnitudes follow the multipole hierarchy $H_T^{u+d} \\ll \\bar E_T^{u-d} \\ll E_T^{u+d}$, $\\tilde E_T$ vanishes at this accuracy, and $\\bar E_T^{u-d}$ at $\\xi=1/3$, $t=-1.02$ GeV$^2$ is in good agreement with lattice QCD in the negative-$x$ region.","pith_inferences":["If the polynomiality proof extends to all orders in the gradient expansion, the mean-field GPD basis could be used to constrain the higher $m\\ge 2$ moments of chiral-odd GPDs that lattice QCD currently does not reach, once the discrete-level contribution is computed.","The quadrupole spin operator $Q^{kl}$, which vanishes in the nucleon matrix element but is generated by the NLO spin-flavor algebra, would become the leading structure in $N\\to\\Delta$ transition GPDs or $\\Delta$-baryon GPDs; the same mean-field formalism could predict those without new model input.","The negative-$x$ agreement with lattice QCD suggests that the ERBL region is dominated by the Dirac-sea (pion-cloud) dynamics captured by the gradient expansion, while the positive-$x$ valence region is governed by the discrete-level contribution; a full single-particle computation would directly separate these two mechanisms.","The hierarchy $H_T^{u+d} \\ll \\bar E_T^{u-d} \\ll E_T^{u+d}$, if confirmed by lattice data, would mean that the multipole order of the mean-field GPD is the organizing principle for the strength of transversity distributions."],"forward_implications":["The complete NLO chiral-odd GPD set in both flavor sectors can be used directly as model input for exclusive pseudoscalar meson production amplitudes, where chiral-odd GPDs couple to pion distribution amplitudes.","The proven polynomiality and first-moment sum rules guarantee that any mean-field-based extraction of tensor charges, $\\kappa_T$, and generalized tensor form factors is internally consistent.","The large-$N_c$ relation $2\\tilde H_T^{u-d} = -E_T^{u-d}$ and the $N_c$-scaling hierarchy give sharp, testable predictions that lattice QCD or other quark models can confirm or refute.","The vanishing of $\\tilde E_T$ at UV-divergent accuracy implies that a nonzero $\\tilde E_T$, if observed, is a direct signal of the UV-finite and discrete-level contributions that this truncation drops.","The peak-shaped ERBL-region GPDs with smooth crossover at $x=\\pm\\xi$ provide a concrete benchmark for upcoming exclusive-production experiments at 12 GeV electron facilities."],"supporting_citations":[{"why":"Supplies the leading-order chiral-odd GPD framework and spin-flavor structures that this work extends to NLO.","marker":"[1]"},{"why":"Establishes the chiral soliton and mean-field description of the nucleon used throughout.","marker":"[24]"},{"why":"Provides the zero-mode quantization, moment of inertia, and single-particle representation methods.","marker":"[26]"},{"why":"Provides the lattice QCD results for $\\bar E_T^{u-d}$ used in the Fig. 6 comparison.","marker":"[52]"},{"why":"Sets the large-$N_c$ kinematics, the $x=O(N_c^{-1})$ domain, and the nucleon mass and parameter values used in numerics.","marker":"[59]"},{"why":"Justifies neglecting the gauge link and gluon contributions to the leading-twist effective operator.","marker":"[61]"},{"why":"Supplies the derivation of moments via integration by parts and the symmetry properties used in the polynomiality proof.","marker":"[70]"}],"fun_headline_variants":["NLO chiral-odd GPDs pass polynomiality and match lattice","Large-Nc GPDs at NLO: multipole structure and lattice fit","Next-to-leading order chiral-odd GPDs: sum rules and lattice agreement","NLO corrections complete chiral-odd GPDs in large-Nc QCD","Chiral-odd GPDs at NLO: multipole mean-field and lattice match"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The numerical predictions and the lattice comparison assume that the UV-divergent Dirac-sea terms kept in the gradient expansion dominate the omitted UV-finite terms and the discrete-level connected-diagram contribution; if those omitted pieces are not small, the computed curves and the claimed agreement with lattice QCD would shift.","fun_headline_variants_meta":{"raw":{"variants":["NLO chiral-odd GPDs pass polynomiality and match lattice","Large-Nc GPDs at NLO: multipole structure and lattice fit","Next-to-leading order chiral-odd GPDs: sum rules and lattice agreement","NLO corrections complete chiral-odd GPDs in large-Nc QCD","Chiral-odd GPDs at NLO: multipole mean-field and lattice match"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000277,"raw_usage":{"total_tokens":1783,"prompt_tokens":1213,"completion_tokens":570,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":829,"completion_tokens_details":{"reasoning_tokens":464}},"tokens_in":829,"tokens_out":570,"duration_ms":5814,"temperature":1.0,"reasoning_tokens":464,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-06T22:36:12.633089+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"At $\\xi=1/3$ and $t=-1.02$ GeV$^2$, evaluate $\\bar E_T^{u-d}(x)$ using the complete sum over all single-particle states including the discrete level; if the positive-$x$ result does not move appreciably toward the lattice points while negative-$x$ agreement is preserved, the truncation to UV-divergent gradient terms is not the dominant approximation.","supporting_citations":[{"cited_title":"Hagler, Phys","cited_arxiv_id":null,"evidence_quote":"Justifies neglecting the gauge link and gluon contributions to the leading-twist effective operator."},{"cited_title":"Twist-4 contribution to unpolarized structure functions F_L and F_2 from instantons","cited_arxiv_id":"hep-ph/9906444","evidence_quote":"Supplies the derivation of moments via integration by parts and the symmetry properties used in the polynomiality proof."}],"review_version":1}