{"id":"1b80913c-8df7-4cf6-beff-371e719b7a26","arxiv_id":"hep-ph/9903358","paper_version":2,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":5.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":5,"one_line_summary":"The Golec-Biernat-Wusthoff saturation model, with parameters fixed from inclusive HERA data, reproduces diffractive deep inelastic scattering and yields a nearly constant diffractive-to-inclusive cross section ratio.","lead":"A previously fitted model of parton saturation inside the proton is applied, with no new fits to the diffractive data, to diffractive deep inelastic scattering measured at HERA. It reproduces the diffractive structure function across beta and x_P and predicts a nearly constant ratio of diffractive to inclusive cross sections.","discovery_kind":"unclear","skeptic_critique":{"model":"selfhosted/deepseek-v4.1-flash","headline":"The load-bearing claim is 'no further parameters' (abstract:s1), but the diffractive normalization is set by a hand-lowered B_D and the small-β qqbar g term carries an overall factor the authors themselves revised by 2 in footnotes 3 and 4.","rationale":"The reader's verdict (CONDITIONAL) and its weakest_assumption point at the right region: the qqbar g term, Eq. (23), dominates at small β and is derived under a leading-log approximation whose error is not quantified. I agree that this is where the paper is softest, but I would sharpen the concern from 'the approximation's error is unquantified' to a concrete, self-flagged normalization issue. The authors' own footnotes 3 and 4 record that the overall factor of the qqbar g contribution was revised by factors of 2 across versions, and the diffractive normalization is additionally rescaled by hand through B_D = 6 rather than the measured 7.1 GeV⁻². These are O(1) effects on exactly the quantity the paper claims as a parameter-free prediction, and the visual comparisons in Figs. 7 and 9 cannot discriminate them. This is a correctness risk, not a circularity: the factors are derived, not fitted, so the claim is not internally inconsistent — it is simply not yet shown to be robust to its own corrections. The reader's other two reservations (no χ² or uncertainty band; the constant ratio being partly structural) are fair but weaker: a model that predicts a constant ratio is making a falsifiable statement that differs from triple Regge (x^-λ), so 'structural' is not by itself a defect. I therefore keep the verdict at CONDITIONAL and do not move it, but I would make the small-β normalization robustness the explicit condition for acceptance.","tokens_in":21103,"tokens_out":4613,"duration_ms":82646,"concrete_test":"Recompute x_P F2^D at the ZEUS kinematics of Fig. 7 (x_P = 0.0042, β ≈ 0.05–0.2, Q² = 8–60 GeV²) twice: once with B_D = 7.1 GeV⁻² (the measured value) instead of 6, and once with the qqbar g overall factor as it stood before the corrections in footnotes 3 and 4. Report the χ² against the ZEUS and H1 points for each variant. If either variant moves the curve outside the quoted data errors at small β, the parameter-free normalization claim (abstract:s1) and the small-β agreement are not established.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central claim is that the model, with parameters fixed from inclusive F2, predicts the diffractive structure function with no new parameters (abstract:s1). Two features undercut the 'parameter-free' reading at the level of normalization, which is what the diffractive prediction most directly tests. (i) The overall diffractive normalization scales as σ0²/(8π B_D) (Eqs. 8–9, 20–21). B_D is not predicted but taken from measurement, and the authors lower it from the measured 7.1 to 6 GeV⁻² (Section 5) on the basis of an unquantified double-dissociation estimate. That is roughly an 18% upward rescaling of the predicted normalization, applied by hand. (ii) The qqbar g term, Eq. (23), dominates at small β — precisely where the authors concede agreement is weakest ('especially in the region of moderate and large values of β', Section 5) — and its overall normalization has been revised by explicit factors of 2: footnote 3 adds a factor of 2 relative to Ref. [13], and footnote 4 inserts a missing 1/2 in the journal version of Eq. (39). A factor-of-2 amplitude error is a factor-of-4 cross-section error, far larger than the visual agreement in Figs. 7 and 9 can discriminate. So the small-β agreement rests on a normalization that is (a) hand-adjusted via B_D and (b) recently corrected by O(1) factors. This is a correctness risk rather than a circularity: the factors are derived, not tuned, but their history means the agreement should be shown to be robust to them.","agreement_with_reader":"partial"},"referee_report":{"model":"selfhosted/deepseek-v4.1-flash","summary":"The manuscript applies the saturation dipole model of the authors' earlier inclusive analysis (Ref. [1]) to diffractive deep inelastic scattering at HERA. The photon fluctuates into a quark-antiquark (or quark-antiquark-gluon) system whose interaction with the proton is given by the saturated dipole cross section sigma-hat = sigma_0 {1 - exp[-r^2/(4 R_0^2(x))]}, with R_0(x) = (1 GeV)(x/x_0)^{lambda/2} and sigma_0, lambda, x_0 fixed from inclusive F2. Section 4 writes the diffractive structure function as three components: transverse quark-antiquark (Eq. 20), longitudinal quark-antiquark (Eq. 21), and quark-antiquark-gluon (Eq. 23), the last in a leading-log(Q^2), strongly ordered approximation. Section 5 compares with ZEUS and H1 data on the beta, x_Pomeron and Q^2 dependence, on the effective Pomeron intercept, and on the diffractive-to-inclusive ratio, using B_D = 6 GeV^-2 and alpha_s = 0.2. Section 6 recasts the results in impact-parameter space, and the Appendix derives the quark-antiquark-gluon amplitude. The central claims are that no new parameters are needed and that saturation naturally yields a nearly constant ratio of diffractive to inclusive cross sections.","tokens_in":21512,"tokens_out":12114,"duration_ms":221255,"significance":"If it holds, the result is a genuine prediction of the diffractive normalization from a dipole cross section fixed entirely on inclusive data, which is a strong statement that one density-dependent object controls both inclusive and diffractive small-x scattering. The near-constant ratio of diffractive to inclusive cross sections is a falsifiable structural prediction that distinguishes the saturation picture from triple-Regge expectations, and the impact-parameter reformulation together with the Appendix derivation is useful reference material. Credit is due for the transparency about the leading-log approximation and about the phenomenological character of the model, and for the breadth of the data comparison (beta, x_Pomeron, Q^2, effective intercept, and the cross-section ratio).","major_comments":[{"comment":"The abstract states that 'No further parameters are needed', but the diffractive normalization is proportional to sigma_0^2/(8 pi B_D) (Eqs. 8-9 and 20-21), and B_D is an external input rather than a prediction. In Section 5 the authors take B_D = 6 GeV^-2, 'somewhat lower' than the measured 7.1 GeV^-2 [21], justified only qualitatively by an unquantified double-dissociation estimate; this raises the predicted normalization by about 18%. The fixed alpha_s = 0.2 enters the quark-antiquark-gluon component linearly (Eq. 23). The claim is therefore better stated as 'no new fitted parameters, with the overall diffractive normalization fixed by the measured B_D'. Please qualify the abstract accordingly and state explicitly the normalization uncertainty implied by the B_D choice (6 versus 7.1 GeV^-2) and by alpha_s.","section":"Abstract; Section 5","authors_acknowledge":null,"authors_acknowledge_p":null},{"comment":"The conclusion rests on visual comparison: Figs. 7-12 show curves against H1 and ZEUS points but no chi-squared, no error treatment, and no statement of which data or which normalization the comparison is most sensitive to. Because the normalization is precisely the quantity the model predicts, a quantitative statement is needed to support 'successfully describe' (abstract:s0). A table of chi-squared, or at least a pull distribution, per data set and per beta region, with the normalization uncertainty from B_D folded in, would make the claim checkable. This is a request for quantification, not an assertion that the comparison is wrong.","section":"Section 5, Figs. 7-12","authors_acknowledge":null,"authors_acknowledge_p":null},{"comment":"The quark-antiquark-gluon term dominates at small beta, which is where the agreement with ZEUS and H1 is weakest; the authors themselves note good agreement 'especially in the region of moderate and large values of beta'. This term is derived in a leading-log(Q^2), strongly ordered approximation, and the authors acknowledge in Section 4 that 'the true kinematical constraints are not exactly fulfilled' and that the approximation 'introduces uncertainties besides those related to the choice of alpha_s'. What remains unresolved is the size of that uncertainty: the normalization of this component has been revised by explicit factors of 2 relative to Ref. [13] (footnote 3), and a missing factor of 1/2 was inserted in Eq. (39) (footnote 4). Please state the estimated residual normalization uncertainty of Eq. (23) and, if feasible, show that the small-beta comparison is stable under a factor-of-2 variation of that component. The acknowledgement is appropriate; the missing piece is quantification.","section":"Section 4, Eq. (23); Section 5, Figs. 7 and 9","authors_acknowledge":null,"authors_acknowledge_p":null}],"minor_comments":[{"comment":"Eq. (19) gives xG approximately equal to (3/(4 pi^2 alpha_s)) sigma_0 R_0^2(x) for Q^2 R_0^2 much greater than 1, but carrying out the integral of Eq. (18) gives (3/(4 pi^2 alpha_s)) sigma_0 / R_0^2(x); Section 2 itself states that 'the gluon density G(x) is proportional to 1/R_0^2(x)'. The displayed prefactor in Eq. (19) appears to be inverted. This does not affect the numerical results, which use Eq. (18) directly, but it should be corrected.","section":"Eq. (19) and Section 2","authors_acknowledge":null,"authors_acknowledge_p":null},{"comment":"The near-constancy of the diffractive-to-inclusive ratio follows from the shared R_0^2(x) scaling of both cross sections (Eqs. 13-14), so Fig. 12 is a consistency check of the model's structure rather than an independent prediction. The text says as much in Section 5 ('a particular feature of our saturation model'), but one explicit sentence at Fig. 12 would prevent the reader from over-reading it.","section":"Eq. (14) and Fig. 12","authors_acknowledge":null,"authors_acknowledge_p":null},{"comment":"Section 5 relates the logarithmic x_Pomeron slope n of x_Pomeron F^D to the effective intercept through n = 1 - 2 alpha_Pomeron(eff). With the standard flux normalization x_Pomeron^{1-2 alpha} this relation does not obviously reproduce the values shown in Fig. 11; please state the convention used for n and for alpha_Pomeron(eff) so that the relation can be checked.","section":"Section 5, effective Pomeron intercept","authors_acknowledge":null,"authors_acknowledge_p":null},{"comment":"Section 5 uses three light flavours only. At the smallest x and beta, charm could be a non-negligible part of the diffractive cross section, and the paper defers this to Ref. [32]. A sentence estimating the size of the neglected charm contribution would help the reader judge the comparison at small beta.","section":"Section 5, flavour content","authors_acknowledge":null,"authors_acknowledge_p":null},{"comment":"Eqs. (11)-(13) are order-of-magnitude estimates obtained with the Theta-function approximation of Eq. (10), and their numerical coefficients differ from a careful treatment by factors of order unity. Since Eq. (14) is described as 'similar to the exact result', the reader should be told explicitly that these are parametric estimates rather than accurate formulae.","section":"Eqs. (11)-(13)","authors_acknowledge":null,"authors_acknowledge_p":null},{"comment":"Refs. [22], [25] and [32] are cited as 'to be published' or 'in preparation'. None is load-bearing for the central claim, but the comparison to [22] in Section 5 would be easier to assess if the relevant result were stated in the text rather than left to the unpublished reference.","section":"References","authors_acknowledge":null,"authors_acknowledge_p":null},{"comment":"There are several typographical errors: 'seperation' and 'mulitple' in Section 2, 'This grows is eventually tamed' in Section 2, 'diffractve' in Section 3, and 'Fig. 40' in the Appendix, which should read Fig. 4. The brace in Eq. (2) separating the transverse and longitudinal wave functions is garbled in the present version and should be reformatted.","section":"Presentation","authors_acknowledge":null,"authors_acknowledge_p":null}],"recommendation":"minor_revision","confidential_remarks":"This is a v2 whose footnotes record corrections of factors of 2 relative to the published version (footnote 3 for Eq. 23, footnote 4 for Eq. 39). The journal should ensure that the corrected normalizations are propagated consistently through the numerical results shown in Figs. 7-12, since the small-beta comparison depends on them. The physics is sound and the paper is a useful contribution; the requested changes are qualifications of the abstract claim and added quantification, not new calculation."},"author_rebuttal":null,"desk_editor":{"model":"selfhosted/deepseek-v4.1-flash","letter":"Bottom line: this is the diffractive companion to the Golec-Biernat–Wüsthoff inclusive saturation fit, and it holds up. The three dipole parameters are fixed on inclusive F2; the diffractive normalization then comes out without a diffractive fit. That is the real content, and it is worth your time. The reader's CONDITIONAL verdict is about right, though I'd weight the framework more positively than the score suggests — the stress-test's factor-of-2 worry is a correction to a cited prior derivation, not a tuning knob, and the authors are upfront about the structural parts of their claim. What's new: the explicit qqbar and qqbar-g momentum-space formulae (Eqs. 20, 21, 23), the impact-parameter reformulation (Eqs. 30–39), and the confrontation with H1 and ZEUS across beta, x_P, Q^2, the effective intercept, and the diffractive-to-inclusive ratio. The three-component decomposition — qqbar-g at small beta, transverse qqbar in the middle, longitudinal at large beta — is wave-function physics and the authors say so; it is a useful organizing picture regardless of the model. Soft spots, in order of how much they matter. The 'no further parameters' line in the abstract is a bit generous. B_D is an input, and they lower it from the measured 7.1 to 6 GeV^-2 on an unquantified double-dissociation estimate — roughly an 18% upward shift in the predicted normalization. alpha_s is pinned at 0.2. Neither is fitted to the diffractive shape, so the shape prediction stands, but the normalization claim is softer than the abstract implies. The comparison is visual throughout: no chi-squared, no error treatment, no theory band. For a paper whose selling point is quantitative agreement, that is the biggest gap. Eq. (23) is leading-log with strong transverse ordering, and the authors flag it themselves ('the true kinematical constraints are not exactly fulfilled'). It dominates at small beta, where the agreement is weakest. The factor-of-2 history in footnotes 3 and 4 is a correction to a prior derivation, not a knob, but it does mean the small-beta normalization deserves an independent cross-check. The constant ratio is structural — Eq. (14) drops out of inserting the saturated ansatz — and the authors present it that way. Read it as a consistency check, not a prediction. Who it's for: anyone working on small-x dipole models, diffraction, or as a baseline for non-linear evolution. It is a standard reference for a reason. Recommendation: send it out. The physics is sound and the parameter-free diffractive normalization is a real result; a referee should push on the error treatment and the B_D choice, not on the framework.","headline":"The diffractive companion to the Golec-Biernat–Wüsthoff saturation fit: the parameter-free diffractive normalization is a real result, but the agreement is shown only by eye and the 'no further parameters' line is softer than the abstract suggests.","tokens_in":22206,"tokens_out":1831,"would_cite":true,"duration_ms":32465,"reading_group":"maybe","serious_thinker":"yes","would_accept_peer_review":true},"pith_extraction":{"msc":[],"pacs":["12.38.-t","13.60.Hb","12.38.Bx"],"model":"selfhosted/deepseek-v4.1-flash","headline":"The same dipole cross section that fits inclusive small-$x$ scattering, with its three parameters already fixed, also accounts for HERA's diffractive deep inelastic data without any new fit — and it predicts that the diffractive and the dif","keywords":["saturation","dipole cross section","diffractive deep inelastic scattering","small x","Pomeron intercept","color transparency","HERA"],"falsifier":"A measurement showing that the diffractive-to-inclusive ratio grows with $1/x$ roughly as $x^{-lambda}$ instead of staying flat — or a diffractive normalization that requires a separate fit rather than following from the inclusive parameters — would contradict the central claim.","tokens_in":22373,"feed_emoji":"⚛️","tokens_out":2504,"duration_ms":39947,"temperature":0.7,"pith_summary":"Deep inelastic scattering at small $x$ can be pictured as the virtual photon splitting into a quark–antiquark pair well before it reaches the proton, with the pair then scattering off the proton as a color dipole. In an earlier paper the authors proposed a dipole cross section $hat{sigma}(x,r^2) = sigma_0[1-exp(-r^2/4R_0^2(x))]$ that stops growing once the dipole separation $r$ exceeds a radius $R_0(x)$ which shrinks as $x$ falls; the three numbers in it were fixed on inclusive $F_2$ data alone. This paper takes that same object, unchanged, and computes diffractive deep inelastic scattering, where the proton stays intact and a large rapidity gap appears. The claim is that the diffractive data from H1 and ZEUS are described at the right normalization, and that the ratio of diffractive to inclusive cross sections comes out roughly constant in $x$ and $Q^2$ rather than rising like a triple-Regge prediction. If that holds, one density-dependent object controls both channels, and diffraction stops looking like a separate soft Pomeron exchange and starts looking like the same saturation physics seen through a different window.","feed_headline":"One fitted dipole cross section predicts HERA diffraction","feed_subtitle":"Three parameters fixed on inclusive data also set the diffractive normalization, with no new fits.","key_machinery":"The saturating dipole cross section $hat{sigma}(x,r^2) = sigma_0{1-exp[-r^2/(4R_0^2(x))]}$ with $R_0(x) = (x/x_0)^{lambda/2}$ GeV$^{-1}$, together with its Fourier-conjugate unintegrated gluon distribution $alpha_s F(x,l_t^2) = (3sigma_0/4pi^2) R_0^2(x), l_t^2 e^{-R_0^2(x)l_t^2}$. The Gaussian form in $l_t^2$ is what makes every $l_t$ integral in the diffractive formulae converge without an external cutoff, and it is the same exponent that supplies the $e^{-Q^2R_0^2}$ factor in the gluon distribution $xG(x,Q^2)$. The diffractive cross section is then built from $hat{sigma}^2$ at zero momentum transfer, divided by the slope $B_D$, with the $t$-dependence assumed exponential.","core_discovery":"The authors argue that diffractive deep inelastic scattering is dominated by the saturation region of the dipole cross section, while inclusive scattering at large $Q^2$ is dominated by the color-transparency region $r sim 2/Q$. Because the diffractive integrand peaks at $r sim 2R_0(x)$, where $hat{sigma}$ has already flattened to $sigma_0$, the diffractive cross section is cut off by the saturation scale itself rather than by an arbitrary infrared cutoff. Working through the small-$r$ approximation of the photon wave function, they obtain $sigma^D/sigma approx (sigma_0/8pi B_D), 1/ln[R_0^2(x)Q^2]$, i.e. an approximately constant ratio, in contrast to the $x^{-2lambda}$ behavior that a non-saturated $hat{sigma} propto r^2$ ansatz would give. Numerically they evaluate three contributions — transverse $qbar q$, longitudinal $qbar q$, and $qbar q g$ — using the parameters $sigma_0 = 23.03$ mb, $lambda = 0.288$, $x_0 = 3.04 imes 10^{-4}$ from the inclusive fit, with $B_D = 6$ GeV$^{-2}$ taken from HERA and $alpha_s = 0.2$ fixed, and compare to the H1 and ZEUS $x_{I P}F_2^D$ data in $beta$, in $x_{I P}$, and through the effective Pomeron intercept. They report agreement, in particular at moderate and large $beta$, and note that the relative size of the three contributions is set by QCD color factors rather than fitted.","pith_inferences":["If the diffractive normalization really is fixed by the inclusive fit, then a future measurement of the diffractive-to-inclusive ratio at much smaller $x$ or higher $Q^2$ is a sharper test of saturation than $F_2$ itself, since the two channels weight different regions of the same dipole cross section.","The prediction that the average transverse momentum of the diffractive final state grows as $x$ falls is separable from the ratio test: an explicit measurement of the $k_t$ spectrum in diffraction would probe the mechanism rather than just the normalization.","The suppression of large dipoles by saturation suggests the same blackening argument should apply to diffractive vector-meson production and to single diffraction in hadron–hadron scattering, where the profile function in central collisions is the analogous quantity.","The authors leave diffractive charm production for later work; since charm masses change the wave-function weighting toward smaller $r$, a charm diffractive measurement would test whether the same $R_0(x)$ sets the scale for heavy quarks too."],"forward_implications":["The diffractive structure function's normalization is a prediction of the inclusive fit, so a discrepancy in overall size would be a failure of the model rather than a free parameter to absorb.","The ratio $sigma^D/sigma$ stays nearly flat in $W$ and $Q^2$ across the measured range, with only a mild rise at small diffractive mass from the longitudinal, higher-twist piece.","The effective Pomeron intercept extracted from the $x_{I P}$ slope rises with $Q^2$ and depends on the diffractive mass, tracking the longitudinal contribution at low mass and being pulled down at small $beta$ by the $ln(Q^2/k^2)$ in the $qbar q g$ term.","Soft contributions to diffraction are suppressed relative to hard ones, so a large fraction of the diffractive cross section comes from $r < 2R_0$ — the basis for calling diffractive deep inelastic scattering a semi-hard process.","Because the gluon distribution implied by the model scales rather than evolves, the $Q^2$ dependence of the data is claimed to be accounted for without adding DGLAP evolution, though the authors note evolution could be added and would give moderate improvement."],"supporting_citations":[{"why":"The earlier inclusive analysis that fixed $sigma_0$, $lambda$ and $x_0$; every number used here is inherited from it, making it the load-bearing reference.","marker":"[1]"},{"why":"H1 diffractive structure function data at fixed $x_{I P} = 0.003$, one of the two datasets the model is compared against.","marker":"[7]"},{"why":"ZEUS diffractive data at $x_{I P} = 0.0042$ used for the $beta$ distributions and for the diffractive-to-inclusive ratio comparison.","marker":"[8]"},{"why":"Supplies the dipole factorization formula relating $gamma^* p$ cross sections to the dipole cross section, the starting point of the whole construction.","marker":"[9]"},{"why":"Provides the unintegrated-gluon relation for $hat{sigma}$ used to convert the momentum-space diffractive formulae into dipole form.","marker":"[10]"},{"why":"Source of the $qbar q g$ diffractive contribution and the gluon-dipole wave function reproduced in Eqs. (34) and (39).","marker":"[13]"},{"why":"ZEUS measurement of the diffractive slope, the origin of the $B_D$ value used to normalize the $t$-integrated cross sections.","marker":"[21]"},{"why":"Independent derivation of the $qbar q g$ triple-Regge limit against which the small-$beta$ behavior of Eq. (23) is checked.","marker":"[12]"},{"why":"The thesis from which the appendix's diagram-by-diagram derivation of the $qbar q g$ amplitude is recalled.","marker":"[29]"}],"fun_headline_variants":["Three-parameter saturation fit describes HERA diffraction","Same dipole fit covers inclusive and diffractive HERA data","Saturation cuts off diffraction where the dipole cross section flattens","Color factors, not fits, set the three diffractive pieces","Diffractive ratio stays constant as saturation cuts off the integrand"],"cache_read_input_tokens":0,"weakest_assumption_plain":"The small-$beta$ part of the prediction rests on the $qbar q g$ term computed in a leading-log$(Q^2)$ approximation with strongly ordered transverse momenta, an approximation the authors themselves say does not exactly satisfy the kinematic constraints and whose error is not quantified.","fun_headline_variants_meta":{"raw":{"variants":["Three-parameter saturation fit describes HERA diffraction","Same dipole fit covers inclusive and diffractive HERA data","Saturation cuts off diffraction where the dipole cross section flattens","Color factors, not fits, set the three diffractive pieces","Diffractive ratio stays constant as saturation cuts off the integrand"]},"model":"selfhosted/deepseek-v4.1-flash","effort":"low","attempts":1,"cost_usd":0.0,"raw_usage":{"total_tokens":2198,"prompt_tokens":968,"completion_tokens":1230,"prompt_tokens_details":null,"completion_tokens_details":{"reasoning_tokens":1160}},"tokens_in":968,"tokens_out":1230,"duration_ms":19228,"image_pages":[],"temperature":1.0,"reasoning_budget":50,"reasoning_tokens":1160,"transport_provider":"selfhosted","cache_read_input_tokens":0,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-10-29T18:33:49.913216+00:00","model_set":{"reader":"selfhosted/deepseek-v4.1-flash"},"falsifier":"A measurement showing that the diffractive-to-inclusive ratio grows with $1/x$ roughly as $x^{-lambda}$ instead of staying flat — or a diffractive normalization that requires a separate fit rather than following from the inclusive parameters — would contradict the central claim.","supporting_citations":[{"cited_title":"Golec-Biernat, M","cited_arxiv_id":null,"evidence_quote":"The earlier inclusive analysis that fixed $sigma_0$, $lambda$ and $x_0$; every number used here is inherited from it, making it the load-bearing reference."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"H1 diffractive structure function data at fixed $x_{I P} = 0.003$, one of the two datasets the model is compared against."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"ZEUS diffractive data at $x_{I P} = 0.0042$ used for the $beta$ distributions and for the diffractive-to-inclusive ratio comparison."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Supplies the dipole factorization formula relating $gamma^* p$ cross sections to the dipole cross section, the starting point of the whole construction."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Provides the unintegrated-gluon relation for $hat{sigma}$ used to convert the momentum-space diffractive formulae into dipole form."},{"cited_title":"W¨ usthoff, Phys","cited_arxiv_id":null,"evidence_quote":"Source of the $qbar q g$ diffractive contribution and the gluon-dipole wave function reproduced in Eqs. (34) and (39)."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"ZEUS measurement of the diffractive slope, the origin of the $B_D$ value used to normalize the $t$-integrated cross sections."},{"cited_title":"Quark-Antiquark-Gluon Jets in DIS Diffractive Dissociation","cited_arxiv_id":"hep-ph/9903265","evidence_quote":"Independent derivation of the $qbar q g$ triple-Regge limit against which the small-$beta$ behavior of Eq. (23) is checked."},{"cited_title":"W¨ usthoff, PhD-thesis, preprint DESY-95-166","cited_arxiv_id":null,"evidence_quote":"The thesis from which the appendix's diagram-by-diagram derivation of the $qbar q g$ amplitude is recalled."}],"review_version":1}