{"id":"97640518-f5c2-478e-b1c7-a847fbf0c492","arxiv_id":"2505.05963","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":5.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":4,"one_line_summary":"Spectator-model estimates of the dihadron beam spin asymmetry reproduce the z and Mh dependence of CLAS data, miss the x and Q2 dependence, and predict measurable asymmetries at future facilities.","lead":"This paper calculates the beam spin asymmetry for pion-pair production in deep inelastic scattering using spectator-model inputs for quark distributions and fragmentation functions. It compares the result with CLAS and CLAS12 data, finds partial agreement, and predicts the asymmetry for COMPASS, the EIC, and EicC.","discovery_kind":"new_application","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The claim that the f1⊗G~ term is negligible and that the Wandzura-Wilczek approximation holds rests entirely on the spectator-model value of G~_ot computed with m=0; a finite quark mass could alter this conclusion.","rationale":"The reader's weakest_assumption identifies the spectator-model calculation of G~_ot and the setting m=0 as the most fragile input. My analysis agrees and sharpens the point: the m=0 choice is not a harmless simplification but is the reason the Im(F_s^*F_p) term disappears, leaving only the Re term to produce G~_ot. Since the paper's WW-supporting statement and its predictions for COMPASS, EIC, and EicC all rely on the smallness of f1⊗G~, this is the single most load-bearing concern. The paper is otherwise a competent, mostly data-independent model estimate: it uses published spectator-model inputs, states its parameter choices clearly, and does not overclaim agreement with CLAS12 data, which it acknowledges deviates. The DGLAP-evolution assumption for e(x) is also a simplification, but it chiefly affects the x- and Q^2-dependence, where the paper already admits discrepancies; the G~_ot issue, by contrast, underpins the WW conclusion that is advertised as a main result. A concrete numerical test varying m can settle whether the f1⊗G~ term remains negligible outside the special m=0 point. Since the reader's verdict was already CONDITIONAL with this concern identified, my assessment does not change the verdict; it reinforces the need for that condition.","tokens_in":26404,"tokens_out":3536,"duration_ms":38188,"concrete_test":"Recompute the f1⊗G~_ot contribution to A_LU^{sin φ_R} in Eq. (22) using Eq. (34) with the fragmenting quark mass set to a finite value, e.g., m = 0.3 GeV (constituent-like) and m = 0.005 GeV (current-like), keeping all other parameters in Eq. (44) fixed. Compare the resulting dash-dotted curves in Figs. 5 and 6 to the e⊗H_1^angle contribution. If the f1⊗G~ term exceeds roughly 10% of the e⊗H_1^angle term or changes sign for either finite m, the claimed WW support is an artifact of the m=0 choice. If it remains negligible for both finite masses, the concern is weakened. A complementary check would be to evaluate G~_ot with an independent model, such as the NJL approach, and compare the size and sign of the f1⊗G~ contribution.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central physics message of the paper is that A_LU^{sin φ_R} is dominated by e(x)⊗H_1^angle and that f1(x)⊗G~^angle is negligible, which is taken as support for the Wandzura-Wilczek approximation. That conclusion depends on the model calculation of the twist-3 DiFF G~_ot in Eqs. (33)-(34). In Eq. (34), the term proportional to Im(F_s^*F_p) is multiplied by the coefficient C, which is explicitly proportional to the fragmenting quark mass m (Eq. 35). The parameter list in Eq. (44) fixes m = 0.0 GeV, so this entire term vanishes, and G~_ot is generated only by the Re(F_s^*F_p) term, proportional to m_s. Thus the numerical result that f1⊗G~ is 'consistent with zero' is not a model-independent feature of QCD; it is a prediction of one specific term in one specific spectator-model diagram, evaluated at a special value of the quark mass. The quark-gluon-quark correlator in Eq. (18) contains other structures, and the spectator-model replacement of the full final-state sum is a strong assumption that the paper does not cross-check against any independent calculation, sum rule, or lattice constraint. If the true G~_ot had a different magnitude or sign, the second term in Eq. (22), with its explicit 1/z and M_h/M prefactors, would change the z- and M_h-dependence of the asymmetry and the WW conclusion would not follow. This is a model-dependence concern rather than an internal inconsistency, but it is load-bearing for the paper's principal phenomenological claim.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper studies the longitudinal-beam single-spin asymmetry A_LU^{sin phi_R} for pion-pair (dihadron) production in SIDIS off an unpolarized proton. The theoretical basis is the collinear factorization formula Eq. (22), taken from Bacchetta and Radici, in which the asymmetry receives two contributions: a twist-3 PDF e(x) coupled to the twist-2 DiFF H_1^angle, and the unpolarized PDF f_1(x) coupled to a twist-3 DiFF G~^angle. The authors use spectator-model results for e(x), D_1, H_1^angle, and G~^angle, with f_1 from CT10, and evolve e(x) by assuming that its DGLAP kernel is the same as that of f_1. They compare the resulting asymmetry with CLAS and CLAS12 data, report good agreement for the z- and M_h-dependencies while acknowledging discrepancies in x and Q^2, and provide predictions for COMPASS, EIC, and EicC. The main physics conclusions are that the e(x) otimes H_1^angle term dominates, the f_1 otimes G~^angle contribution is consistent with zero and thereby supports the Wandzura-Wilczek approximation, and that QCD evolution of e(x) significantly affects the x- and Q^2-dependence.","tokens_in":26815,"tokens_out":5501,"duration_ms":57844,"significance":"If the results hold, the paper provides a complete model estimate of a twist-3 observable that can access e(x) through dihadron production, and it is apparently the first to include the twist-3 DiFF G~^angle in this asymmetry. The authors give explicit model expressions, parameter tables, and a clear enumeration of the input assumptions, and they produce falsifiable predictions for COMPASS, EIC, and EicC. The comparison with published CLAS and CLAS12 data is a useful benchmark, and the discussion of QCD evolution in a twist-3 PDF analysis is timely. The central numerical and phenomenological conclusions, however, are not yet supported at the level claimed: the agreement with data is qualitative, and the Wandzura-Wilczek conclusion rests on a single spectator-model term with the quark mass set to zero.","major_comments":[{"comment":"The paper's central quantitative claim — \"good agreement\" with the CLAS z- and M_h-dependent asymmetries — is not supported by any numerical goodness-of-fit measure. The model curves have no uncertainty bands, no chi^2 or pull values are given, and the text itself concedes that the x- and Q^2-dependencies disagree. Since the x- and Q^2-shapes are precisely the observables that carry the e(x) input and its DGLAP evolution, a quantitative comparison (for example chi^2/ndf or bin-wise pulls) is required before the claim of agreement can be evaluated. Without this, the comparison in Figs. 5 and 6 is illustrative only and does not establish the model's ability to describe the data.","section":"Sec. III.C, Figs. 5 and 6"},{"comment":"The conclusion that the f_1 otimes G~^angle term is negligible and \"supports the Wandzura-Wilczek approximation\" depends entirely on the spectator-model expression for G~_ot. In Eq. (34) the Im(F_s^* F_p) term is multiplied by the coefficient C, and Eq. (35) shows that C is proportional to the fragmenting quark mass m; with m = 0.0 GeV fixed in Eq. (44), this term vanishes and only the Re(F_s^* F_p) contribution, proportional to m_s, remains. The quark-gluon-quark correlator in Eq. (18) is a general nonperturbative object, and the spectator-model replacement of the full final-state sum is a model assumption rather than a QCD result. If the true G~_ot had a different magnitude or sign, the second term in Eq. (22), with its explicit M_h/(zM) and 1/z prefactors, would change the z- and M_h-dependence of the asymmetry and the Wandzura-Wilczek conclusion would not follow. The manuscript should either provide a cross-check of G~_ot (for example, a nonzero m test, an alternative DiFF model, or a model-independent bound) or soften the Wandzura-Wilczek claim to a model-dependent statement.","section":"Sec. III.B, Eqs. (33)-(35) and (44)"},{"comment":"The assumption that the DGLAP evolution kernel for the twist-3 PDF e(x) is identical to that of f_1 is stated but not justified. Twist-3 distributions generally mix with quark-gluon-quark correlation functions under evolution, and the evolution of e(x) is not in general identical to that of a twist-2 PDF. Because one of the paper's stated conclusions is that QCD evolution \"plays an essential role\" in these observables, the uncertainty introduced by this assumption should be quantified or checked against available LO twist-3 evolution results. As it stands, the x- and Q^2-dependence of the evolved predictions is not fully under control.","section":"Sec. III.A, DGLAP evolution of e(x)"}],"minor_comments":[{"comment":"The notation |M| appears where the nucleon mass M is meant, and the placement of the factors 1/x and |R| in the integrands of Eqs. (48)-(52) should be clarified; please ensure that all kinematic prefactors are defined consistently with Eq. (22).","section":"Eqs. (48)-(52)"},{"comment":"The captions use \"dashed lines\" for two different quantities — the e(x) H_1^angle term and the total asymmetry without QCD evolution — which makes the panels ambiguous. Please use distinct line styles and a clear legend in all three figures.","section":"Captions of Figs. 5-7"},{"comment":"There is an incomplete sentence and a typo in the paragraph introducing the CLAS kinematics: \"...scattered off an unpolarized proton the kinematical region\" should be rewritten, and \"dihadron production production\" should be corrected.","section":"Sec. III.C, text before Eq. (53)"},{"comment":"Several typeset symbols are garbled, for example \"H <)\" should be H^{angle} throughout; please correct the encoding so that the partial-wave expansions are readable.","section":"Sec. II, Eq. (16) and surrounding text"}],"recommendation":"major_revision","confidential_remarks":"The paper is within the journal's scope and the application of the G~^angle DiFF to this asymmetry appears new, so I see no novelty or overlap concern. The main issue is that the phenomenological conclusions are stronger than the quantitative analysis supports: the data comparison lacks a goodness-of-fit measure, and the Wandzura-Wilczek conclusion hinges on one spectator-model term evaluated at m = 0. A revision that adds a quantitative comparison and either cross-checks or softens the WW claim would materially strengthen the paper."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Short version: this is a competent, honest phenomenological estimate of A_LU^sin phi_R in dihadron SIDIS, and it is the first to include the T-odd twist-3 DiFF G~^angle together with DGLAP-evolved e(x). It gives testable predictions for COMPASS, EIC, and EicC. The paper is worth a serious referee, but the Wandzura-Wilczek conclusion should be read with caution.\n\nWhat is actually new: Ref. [28] extracted e(x) from the same observable, but the spectator-model computation of G~^angle_ot, the inclusion of its f1⊗G~ term, and the DGLAP evolution of e(x) through HOPPET have not been combined before. The evolution changes the x- and Q2-dependence measurably, and the predictions for future facilities are new. The presentation is also refreshingly transparent: the authors flag the x- and Q2-disagreements with CLAS/CLAS12 data explicitly rather than claiming a global fit.\n\nThe soft spots, in rough order of importance. First, the claim that f1⊗G~ is negligible and that this supports the Wandzura-Wilczek approximation is not robust. The G~ calculation in Eqs. (33)-(34) is a spectator-model estimate, and the quark mass is fixed to zero, which kills the Im(F_s^*F_p) term; the numerical result then comes entirely from one term in one diagram. If the true quark-gluon-quark correlator differs, the magnitude or sign of G~ could change, and the second term in Eq. (22) has explicit 1/z and M_h/M prefactors that would alter the shape of the asymmetry. The paper does not cross-check this against any independent calculation or sum rule. That said, this is a model-dependence caveat, not an internal inconsistency, and the e⊗H term does dominate in their model. Second, the DGLAP evolution of e(x) is assumed to follow the f1 kernel; that is an assumption, and its effect on the x-dependence is large enough to matter. Third, there are no uncertainty bands from the model parameters, and the agreement with data is qualitative; a referee should ask for at least a sensitivity scan of the spectator parameters.\n\nOverall, I agree with the reader's conditional verdict. This is a useful and honest subfield contribution, not a breakthrough. I would send it to peer review rather than desk reject, with requests for a robustness check on G~ and an explicit statement that the WW inference is model-dependent. If you work on e(x) or dihadron SIDIS, this is worth citing for its predictions and clear treatment of evolution effects.","headline":"A solid, transparent model-based estimate of the dihadron beam-spin asymmetry that deserves serious refereeing, but the Wandzura-Wilczek inference is only as strong as the spectator-model twist-3 DiFF with the quark mass set to zero.","tokens_in":27356,"tokens_out":2680,"would_cite":true,"duration_ms":28851,"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 longitudinal beam single-spin asymmetry in pion-pair SIDIS is dominated by the $e(x)\\otimes H_1^{\\sphericalangle}$ term, with the competing $f_1\\otimes\\widetilde{G}^{\\sphericalangle}$ term negligible.","keywords":["beam single-spin asymmetry","dihadron fragmentation function","twist-3 PDF e(x)","semi-inclusive deep inelastic scattering","Wandzura-Wilczek approximation","spectator model","DGLAP evolution","pion-pair production"],"falsifier":"A dedicated measurement of $A_{LU}^{\\sin\\phi_R}$ at COMPASS, EIC, or EicC that separates the $e\\otimes H_1^{\\sphericalangle}$ and $f_1\\otimes\\widetilde{G}^{\\sphericalangle}$ contributions -- for instance by inspecting the $M_h$-dependence at low $M_h$ (below about 0.63 GeV), where the model's $\\widetilde{G}_{ot}^{\\sphericalangle}$ is largest -- would settle whether the second term is truly negligible. A non-negligible $f_1\\otimes\\widetilde{G}^{\\sphericalangle}$ contribution would falsify the paper's Wandzura-Wilczek conclusion.","tokens_in":26216,"feed_emoji":"⚛️","tokens_out":10593,"duration_ms":90337,"temperature":0.7,"pith_summary":"The paper tries to establish that the longitudinal beam single-spin asymmetry $A_{LU}^{\\sin \\phi_R}$ measured in pion-pair production in semi-inclusive deep inelastic scattering is controlled by a single twist-3 term: the convolution of the quark distribution $e(x)$ with the chiral-odd dihadron fragmentation function $H_1^{\\sphericalangle}$. Using spectator-model inputs for $e(x)$, the unpolarized dihadron fragmentation function $D_1$, the interference function $H_1^{\\sphericalangle}$, and the twist-3 fragmentation function $\\widetilde{G}^{\\sphericalangle}$, the authors reproduce the $z$- and $M_h$-dependence of the CLAS data once DGLAP evolution of $e(x)$ is included. The second contribution, $f_1 \\otimes \\widetilde{G}^{\\sphericalangle}$, comes out negligible in every kinematic region, which the authors take as support for the Wandzura-Wilczek approximation for this observable. If the claim holds, this asymmetry is a clean collinear access route to $e(x)$ -- a twist-3 distribution linked to the nucleon scalar charge and to the proton mass decomposition -- with testable predictions for COMPASS, EIC, and EicC.","feed_headline":"Pion-pair beam asymmetry traces to one twist-3 term","feed_subtitle":"The e(x) term matches CLAS data; the f1-G~ term vanishes, supporting the Wandzura-Wilczek approximation.","key_machinery":"The load-bearing object is the ratio in Eq. (22): a kinematic factor times the charge-weighted sum $[x e(x)H_{1,ot}^{\\sphericalangle}(z,M_h^2) + (M_h/(zM))f_1(x)\\widetilde{G}_{ot}^{\\sphericalangle}(z,M_h^2)]$ divided by the unpolarized sum $f_1(x)D_{1,oo}(z,M_h^2)$. The spectator model supplies each ingredient: $e(x)$ from scalar and axial-vector diquark channels, $D_{1,oo}$ and $H_{1,ot}^{\\sphericalangle}$ from the two-hadron fragmentation correlator, and $\\widetilde{G}_{ot}^{\\sphericalangle}$ from the quark-gluon-quark correlator. With the fragmenting quark mass set to $m=0$, the imaginary part of $F_s^*F_p$ drops out of $\\widetilde{G}_{ot}^{\\sphericalangle}$, leaving only the real-part term; this is the step that makes the $f_1\\otimes\\widetilde{G}^{\\sphericalangle}$ contribution numerically tiny. DGLAP evolution of $e(x)$ (implemented with HOPPET) changes the $x$- and $Q^2$-shape of the asymmetry but barely affects its $z$- and $M_h$-dependence. The Wandzura-Wilczek approximation invoked here is the statement that the genuine quark-gluon-quark piece of the twist-3 function can be neglected, leaving only the twist-2-derived part.","core_discovery":"The central claim is that the beam single-spin asymmetry $A_{LU}^{\\sin \\phi_R}$ for pion-pair production in SIDIS is dominated by the convolution of the twist-3 PDF $e(x)$ with the twist-2 chiral-odd dihadron fragmentation function (DiFF) $H_{1,ot}^{\\sphericalangle}$, while the second term, involving $f_1(x)$ and the twist-3 DiFF $\\widetilde{G}_{ot}^{\\sphericalangle}$, is negligible. With spectator-model inputs for all four functions and DGLAP evolution applied to $e(x)$, Eq. (22) reproduces the $z$- and $M_h$-dependence of the CLAS measurement, although the $x$- and $Q^2$-dependence show discrepancies with both CLAS and CLAS12 data. In every kinematic region studied (CLAS, CLAS12, COMPASS, EIC, EicC), the $f_1 \\otimes \\widetilde{G}^{\\sphericalangle}$ contribution is consistent with zero, which the authors read as evidence for the Wandzura-Wilczek approximation. They conclude that the asymmetry remains measurable at the future facilities, with its magnitude suppressed at higher $Q^2$, and that QCD evolution of $e(x)$ is essential for such twist-3 analyses.","pith_inferences":["A lattice QCD evaluation of the quark-gluon-quark correlator that defines $\\widetilde{G}^{\\sphericalangle}$ would test the key model assumption directly: if that correlator is not negligible at the relevant scales, the vanishing of the $f_1\\otimes\\widetilde{G}^{\\sphericalangle}$ term is a spectator-model artifact rather than a property of QCD.","The same spectator-model machinery could be applied to other twist-3 dihadron observables, such as the asymmetry with a transversely polarized target; a similarly negligible $\\widetilde{G}^{\\sphericalangle}$ there would show the Wandzura-Wilczek approximation extends beyond this one beam-spin observable.","The paper assumes the DGLAP kernel for $e(x)$ is the same as for $f_1(x)$; a comparison of $A_{LU}^{\\sin\\phi_R}$ measurements at two well-separated $Q^2$ values, for example CLAS12 versus EIC, would test that assumption and could reveal a distinct twist-3 evolution kernel.","The near-zero $f_1\\otimes\\widetilde{G}^{\\sphericalangle}$ prediction could itself be checked by isolating the low-$M_h$ region (below about 0.63 GeV) where the real-part term in $\\widetilde{G}_{ot}^{\\sphericalangle}$ is largest; a measurable contribution there would falsify the approximation."],"forward_implications":["Because the $e(x)\\otimes H_1^{\\sphericalangle}$ term alone reproduces the measured $z$- and $M_h$-dependence, this asymmetry is a viable collinear probe of $e(x)$ that does not require a model for the twist-3 fragmentation function.","The $f_1\\otimes\\widetilde{G}^{\\sphericalangle}$ term is reported to be consistent with zero at all studied kinematics, so under the paper's assumptions the Wandzura-Wilczek approximation is valid for this observable and future extractions may drop this term.","DGLAP evolution of $e(x)$ changes the asymmetry by enhancing it at small $x$ and small $Q^2$ and suppressing it at large $x$ and large $Q^2$; analyses that ignore twist-3 evolution would misestimate $e(x)$.","The same $x$-, $z$-, and $M_h$-shapes appear at COMPASS, EIC, and EicC with reduced magnitude, and the asymmetries are large enough to be measured.","The remaining discrepancies in the $x$- and $Q^2$-dependence at CLAS and CLAS12 indicate that the model, the evolution treatment, or the scale dependence of the DiFFs needs refinement at higher scales."],"supporting_citations":[{"why":"Establishes the asymmetry expression in Eq. (22) and the first point-by-point extraction method for e(x) from dihadron SIDIS.","marker":"[28]"},{"why":"Provides the collinear cross-section formalism for dihadron SIDIS, including the definition of the twist-3 DiFF G~.","marker":"[36]"},{"why":"Introduces the partial-wave expansion and the collinear DiFFs D1 and H1 used in the calculation.","marker":"[37]"},{"why":"Spectator-model calculation of the unpolarized and interference DiFFs D1,oo and H1,ot, with the parameter values adopted here.","marker":"[40]"},{"why":"CLAS data for A_LU^{sin phi_R}; the z- and M_h-dependence are the paper's main comparison.","marker":"[47]"},{"why":"CLAS12 data used for the x-, z-, and M_h-dependence comparison.","marker":"[48]"},{"why":"Spectator-model calculation of the twist-3 DiFF G~_ot from the quark-gluon-quark correlator.","marker":"[49]"},{"why":"Spectator-model calculation of e(x) whose Set 2 input is adopted.","marker":"[50]"},{"why":"Provides the NLO CT10 unpolarized PDF f1 used in the asymmetry ratio.","marker":"[53]"},{"why":"HOPPET implementation used for the DGLAP evolution of e(x).","marker":"[58]"}],"fun_headline_variants":["Beam spin asymmetry in pion pairs traced to twist-3 e(x)","Pion-pair beam asymmetry: only twist-3 e(x) matters","Vanishing f1-G~ term supports Wandzura-Wilczek","e(x) beam asymmetry predicted for COMPASS, EIC, EicC"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The paper's conclusion that the second term is negligible depends on a spectator-model calculation of the twist-3 fragmentation function with the fragmenting quark mass set to zero. If that calculation does not capture the real quark-gluon-quark correlation, the neglected term could be sizeable and the Wandzura-Wilczek conclusion would fail.","fun_headline_variants_meta":{"raw":{"variants":["Beam spin asymmetry in pion pairs traced to twist-3 e(x)","Pion-pair beam asymmetry: only twist-3 e(x) matters","Vanishing f1-G~ term supports Wandzura-Wilczek","e(x) beam asymmetry predicted for COMPASS, EIC, EicC"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.00108,"raw_usage":{"total_tokens":4588,"prompt_tokens":1088,"completion_tokens":3500,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":704,"completion_tokens_details":{"reasoning_tokens":3417}},"tokens_in":704,"tokens_out":3500,"duration_ms":26298,"temperature":1.0,"reasoning_tokens":3417,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-15T22:52:10.960310+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"A dedicated measurement of $A_{LU}^{\\sin\\phi_R}$ at COMPASS, EIC, or EicC that separates the $e\\otimes H_1^{\\sphericalangle}$ and $f_1\\otimes\\widetilde{G}^{\\sphericalangle}$ contributions -- for instance by inspecting the $M_h$-dependence at low $M_h$ (below about 0.63 GeV), where the model's $\\widetilde{G}_{ot}^{\\sphericalangle}$ is largest -- would settle whether the second term is truly negligible. A non-negligible $f_1\\otimes\\widetilde{G}^{\\sphericalangle}$ contribution would falsify the paper's Wandzura-Wilczek conclusion.","supporting_citations":[{"cited_title":"Courtoy, A","cited_arxiv_id":null,"evidence_quote":"Establishes the asymmetry expression in Eq. (22) and the first point-by-point extraction method for e(x) from dihadron SIDIS."},{"cited_title":"Radici, R","cited_arxiv_id":null,"evidence_quote":"Provides the collinear cross-section formalism for dihadron SIDIS, including the definition of the twist-3 DiFF G~."},{"cited_title":"Bacchetta and M","cited_arxiv_id":null,"evidence_quote":"Introduces the partial-wave expansion and the collinear DiFFs D1 and H1 used in the calculation."},{"cited_title":"Bianconi, S","cited_arxiv_id":null,"evidence_quote":"Spectator-model calculation of the unpolarized and interference DiFFs D1,oo and H1,ot, with the parameter values adopted here."},{"cited_title":"Diehl, Prog","cited_arxiv_id":null,"evidence_quote":"CLAS data for A_LU^{sin phi_R}; the z- and M_h-dependence are the paper's main comparison."},{"cited_title":"Mirazita et al","cited_arxiv_id":null,"evidence_quote":"CLAS12 data used for the x-, z-, and M_h-dependence comparison."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Spectator-model calculation of the twist-3 DiFF G~_ot from the quark-gluon-quark correlator."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Spectator-model calculation of e(x) whose Set 2 input is adopted."},{"cited_title":"Lu and I","cited_arxiv_id":null,"evidence_quote":"Provides the NLO CT10 unpolarized PDF f1 used in the asymmetry ratio."}],"review_version":1}