{"id":"cf48f3f9-54a4-4fe2-9272-13aaf5693076","arxiv_id":"2504.15002","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":5.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":2,"one_line_summary":"Quark-loop NLO corrections to pure annihilation B decays leave branching ratios nearly unchanged but significantly enhance direct and mixing-induced CP asymmetries, with Bd to phi phi reaching about -40 percent.","lead":"This paper adds next-to-leading-order quark-loop corrections to a set of rare B-meson decays that only proceed through annihilation diagrams. It finds the corrections barely change the decay rates but substantially change the predicted CP asymmetries, giving testable Standard Model benchmarks for LHCb and Belle II.","discovery_kind":"extension","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Neglected NLO corrections could shift CP asymmetries by an amount comparable to the claimed effect; the paper's own Section 3 asserts rather than demonstrates their smallness.","rationale":"The reader's weakest_assumption correctly identifies the load-bearing issue: the paper asserts, without quantitative support, that the omitted NLO corrections are less significant than the quark-loop and chromo-magnetic-penguin contributions. This is the right concern because the paper's own Section 3 admits the omission and the headline observable (A_CP^dir(Bd→φ_L φ_L) going from 0 to -40%) is a small-difference quantity; even a modest omitted imaginary part can shift it substantially. The paper does have independent support: Table 2 shows a consistent relation between |A_NLO|/|A_LO| ≈ 0.36-0.38 and |A_LO+NLO|/|A_LO| ≈ 1.02, the branching ratios in Table 3 are broadly consistent with experiment, and the B→PV exclusion is a stated limitation. But the concern is genuine and testable, so the reader's CONDITIONAL verdict is appropriate rather than ACCEPT. I agree with the reader's framing rather than disagreeing, and I have not found a separate stronger concern; the analytic-formula verification issue noted by the reader is secondary to the quantitative neglect issue. The recommended test is a standard PQCD calculation of the omitted diagrams, which the authors are equipped to perform, so the concern can be settled without new experimental data. The load_bearing_attack field already quotes the Section 3 passage and cites Tables 2 and 4; the concrete_test field proposes one specific computation that would settle it. I therefore recommend CONDITIONAL, consistent with the reader's verdict but emphasizing that the condition is explicit: the omitted NLO corrections must be shown to be numerically small by the same standards used for the kept diagrams.","tokens_in":18247,"tokens_out":4311,"duration_ms":31421,"concrete_test":"Compute the one-gluon virtual corrections to the leading-order annihilation diagrams of Fig. 1 with the same twist-2 and two-particle twist-3 LCDAs and the same Sudakov factors, and add them to the amplitudes before evaluating Table 4. If the resulting A_CP^dir(Bd→φ_L φ_L) changes by more than about 5 percentage points (i.e., away from -40%), the neglected contribution is not negligible and the central claim is not established. Alternatively, an analytic order-of-magnitude bound on the imaginary part of the neglected diagrams relative to |A_LO| would suffice if it shows the bound is below the few-percent level.","verdict_should_be":"CONDITIONAL","load_bearing_attack":"The most load-bearing assumption is that the quark-loop and chromo-magnetic-penguin diagrams of Fig. 2 dominate the NLO corrections, so that the triangle diagrams of Fig. 3 and the QCD corrections to the Fig. 1 leading-order diagrams can be neglected. Section 3 asserts this on the basis of CKM/Wilson-coefficient enhancement and Ref. [47], but gives no quantitative estimate within the present PQCD framework. The headline observable is exactly the kind of quantity that is not robust to this assumption: A_CP^dir(Bd→φ_L φ_L) moves from 0 at LO to about -40% with the NLO amplitude (Table 4), so an omitted amplitude with an imaginary part of order 10-20% of the LO one could shift this prediction by tens of percent. Moreover, the claim that the quark-loop contribution does not affect the transverse/normal VV amplitudes relies on the leading-power argument that internal gluons must be transversely polarized; the same Section 3 admits this argument breaks down at twist-3, and the twist-3 NLO contribution for the transverse amplitudes is not shown to vanish. The B→PV exclusion and the statements in the abstract/conclusions that the NLO 'significantly enhances' A_CP rest on the same cancellation pattern, so a hidden comparable correction would change the central claim rather than just the error bars.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper computes next-to-leading-order quark-loop and chromo-magnetic-penguin contributions to pure-annihilation B-meson decays (Bs→ππ, ρρ, ρω, ωω; Bd→K+K−, K*+K*−, φφ) in the PQCD approach. The main finding is that the NLO amplitudes have large imaginary parts with signs opposite to the leading-order amplitudes, so |A_LO+A_NLO|/|A_LO|≈1.02 and the branching ratios barely change, while the resulting strong phases substantially modify the CP asymmetries. The central numerical results are in Tables 3 and 4, including A_CP^dir(Bd→φ_Lφ_L)≈−40% at NLO compared with 0 at LO. The paper excludes B→PV modes on the grounds that the quark-loop two-gluon form factor does not contribute to those channels.","tokens_in":18487,"tokens_out":11302,"duration_ms":103363,"significance":"If the result holds, it would establish that, within PQCD, the quark-loop and O8g diagrams provide the dominant source of strong phases in pure-annihilation decays, turning several CP asymmetries from unobservably small to measurable values (notably Bd→φφ) while leaving branching ratios essentially unchanged. The predictions are genuine output of the calculation rather than fits: the inputs are the standard PQCD parameters (ω_b, Gegenbauer moments, decay constants) and CKM elements from independent sources, and the amplitude ratios in Table 2 are internally consistent. The main weakness is that the dominance of the selected NLO diagrams is asserted rather than demonstrated, and the quoted uncertainties on A_mix make several of the claimed 'significant enhancements' statistically weak.","major_comments":[{"comment":"The central claim rests on the assertion that the quark-loop and chromo-magnetic-penguin diagrams of Fig. 2 dominate the NLO corrections, while the triangle diagrams of Fig. 3 and the QCD corrections to the Fig. 1 diagrams can be neglected. The text states this is true because of CKM/Wilson enhancement and cites Ref. [47], but no quantitative estimate is given within the PQCD framework. This is load-bearing because the headline observables in Table 4 are generated by the imaginary part of A_NLO; for example, A_CP^dir(Bd→φ_Lφ_L) changes from 0 at LO to −39.7% at NLO. An omitted amplitude whose imaginary part is 10–20% of A_LO would shift these predictions by tens of percent. Please provide an explicit estimate of the leading neglected terms, or state clearly that the predictions assume their smallness.","section":"Section 3, after Eq. (11)"},{"comment":"The abstract and conclusions claim that NLO QCD corrections 'significantly enhance' both A_CP^dir and A_CP^mix. The quoted uncertainties do not support this for A_mix: for Bs→π+π− the LO value 35.9+15.6−11.2 and NLO value −4.2+21.4−9.0 overlap within 1σ, and for Bd→K+K− the LO −47.0+15.7−18.8 and NLO −2.2+19.1−26.4 also overlap within 1σ. Even for Bd→φ_Lφ_L, A_mix=27.8+5.7−25.9 is consistent with 0 at the 1σ level. The only cleanly significant effect in Table 4 is A_CP^dir(Bd→φ_Lφ_L). The wording should be qualified, or the uncertainty propagation should be explained if the asymmetric errors are not intended as 1σ intervals.","section":"Table 4 and Conclusions/Abstract"},{"comment":"The central mechanism is presented through the amplitude ratios for Bs→π+π− and Bs→ρ_Lρ_L, but no analogous decomposition is given for Bd→φ_Lφ_L, the channel with the largest claimed effect. Since A_CP^dir for φφ is zero at LO and becomes −39.7% at NLO, the reader cannot check whether the effect is driven by the same |A_NLO|/|A_LO| ≈ 0.36 cancellation pattern or by a different numerical balance. Please include the amplitude decomposition (or the relevant ratios and strong phases) for Bd→φφ, and ideally for all modes in Table 3.","section":"Table 2 and Table 4"}],"minor_comments":[{"comment":"The sentence 'In Table 2, we also list the experimental results' should read 'Table 3'; the experimental values appear in Table 3, not Table 2.","section":"Section 4, paragraph before Table 3"},{"comment":"The B-meson shape parameters are quoted as ω_b = 0.40±0.5 GeV and ω_b = 0.50±0.5 GeV; as printed the errors exceed the central values. This is presumably a typo for ±0.05 GeV and should be corrected, since ω_b is a major source of the quoted uncertainties.","section":"Section 4, Eq. (21)"},{"comment":"The statement that the twist-3 ⟨P(p2)V(p3)|g*g*⟩ matrix element vanishes because the number of gamma matrices in the pseudo-scalar and vector LCDAs is odd is asserted without showing the trace. Since this is used to justify excluding B→PV modes, a one-line demonstration would help the reader verify the claim.","section":"Section 3, B→PV argument"},{"comment":"The phrase 'evidently revealed' is too emphatic for a calculation with the large uncertainties shown in Table 4; 'indicate' or 'suggest' would be more appropriate.","section":"Abstract and Conclusions"},{"comment":"There is a spelling error in 'diffesr' (should be 'differs') in the discussion of the Lorentz structure of power-suppressed collinear quark fields.","section":"Section 3, first paragraph"}],"recommendation":"major_revision","confidential_remarks":"The manuscript is a direct continuation of the authors' own Ref. [47], and the novelty lies in the PQCD treatment and the extension to vector-meson final states. The main risk is the unquantified dominance assumption in Section 3; I would ask the authors to supply a numerical estimate or a scaling argument before publication. The reliance on Ref. [47] for the B→g*g* form factors is legitimate but should be cross-checked, since it carries the leading-power argument that underlies the B→PV exclusion."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"This is a worthwhile follow-up to Ref. [47], extending the quark-loop enhancement from twist-2 only to twist-3 LCDAs and to longitudinally polarized vector-vector final states. The new numerical output is concrete and testable, especially A_CP^dir(Bd→φ_L φ_L) ≈ -40% where the leading order gives zero. The internal logic is coherent: the NLO amplitude is 35–40% of the LO one, its large imaginary part cancels in the branching ratios (|A_LO+NLO|/|A_LO| ≈ 1.02 in Table 2), and the CP asymmetries come from the strong phases that the loop generates. None of that is fitted to CP data; it follows from CKM inputs and independently sourced LCDA moments. That is genuine progress within the PQCD program, not a repackaging.\n\nThe soft spot is the one the stress-test correctly identifies. Section 3 dismisses the triangle diagrams and QCD corrections to the LO graphs as \"less significant\", citing CKM/Wilson enhancement and Ref. [47], but it gives no order-of-magnitude estimate in the present framework. For a difference-type observable like A_CP^dir, an omitted amplitude with an imaginary part of order 10–20% of the LO amplitude could shift the headline -40% by many points. The same Section also claims the NLO correction only affects longitudinal VV polarizations; that argument is made at leading power, and the text itself admits twist-3 internal gluons need not be transverse. No proof is given that the twist-3 NLO contribution to the transverse VV amplitudes vanishes. These concerns are real but not fatal: the qualitative claim that quark loops generate strong phases and can change CP asymmetries in annihilation modes likely survives, though the precise numbers should be treated as the central values of an incomplete NLO calculation.\n\nMinor but visible: Table 1 quotes ω_b = 0.40 ± 0.5 GeV, which is almost certainly a typo and would inflate uncertainties if taken literally; and the text says experimental results are listed in Table 2 when they are in Table 3. Both are easy fixes.\n\nThe paper deserves a serious referee. A referee should push for a quantitative bound on the neglected NLO pieces, or at least a more careful argument for why they cannot compete with the charm-loop phase. For researchers working on PQCD annihilation decays or B→VV CP asymmetries, the B→φφ benchmark is a concrete target for LHCb and Belle II. I would send it to review and accept conditionally once the neglected-corrections question is addressed.","headline":"A real extension of the quark-loop mechanism to twist-3 and vector-vector final states, capped by a striking -40% direct CP asymmetry benchmark in Bd→φφ, but the paper asserts rather than demonstrates that other NLO corrections are negligible.","tokens_in":19069,"tokens_out":3668,"would_cite":true,"duration_ms":36810,"reading_group":"yes","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":[],"pacs":["13.25.Hw","12.38.Bx","11.30.Er"],"model":"deepseek-v4-flash","headline":"The paper claims that quark-loop next-to-leading-order QCD corrections, though nearly invisible in branching ratios, become the dominant source of strong phases in pure annihilation B decays and sharply enhance both direct and…","keywords":["PQCD approach","pure annihilation B decays","quark-loop corrections","CP asymmetries","strong phases","charmless nonleptonic B decays","chromo-magnetic penguin","light-cone distribution amplitudes"],"falsifier":"Compute the triangle diagrams of Fig. 3 and the full QCD corrections to the Fig. 1 diagrams in the same PQCD framework: if their imaginary parts are comparable to or larger than the quark-loop contribution, the claimed dominance and the predicted strong phases collapse. On the experimental side, measuring $A_{\\rm CP}^{\\rm dir}(\\bar B_d\\to\\phi_L\\phi_L)$ near zero rather than the predicted about $-40\\%$ would falsify the central claim.","tokens_in":17982,"feed_emoji":"⚛️","tokens_out":7604,"duration_ms":67140,"temperature":0.7,"pith_summary":"The paper argues that a specific next-to-leading-order QCD effect—the quark-loop contraction of four-quark operators plus the chromo-magnetic penguin—dominates the NLO corrections to pure annihilation charmless B-meson decays in the PQCD approach. The NLO amplitudes carry large imaginary parts with signs opposite to the leading-order amplitudes, so the two cancel in magnitude and the branching ratios stay close to their leading-order values ($|A_{\\rm LO+NLO}|/|A_{\\rm LO}|\\approx 1.02$). But that same cancellation creates sizable strong-phase differences, so the CP-violating observables change dramatically: for $\\bar B_d\\to\\phi_L\\phi_L$, the direct asymmetry goes from $0$ at leading order to about $-40\\%$, and the mixing-induced asymmetry from $0$ to about $28\\%$. The paper thereby turns pure annihilation decays, which are difficult for collinear factorization because of endpoint divergences, into a concrete probe of strong phases and CP violation.","feed_headline":"Quark loops take B→φφ CP asymmetry from 0 to −40%","feed_subtitle":"The same NLO correction barely moves branching ratios while supplying the strong phases that drive CP violation.","key_machinery":"The machinery is the reduction of the quark-loop diagrams—the gluon attached to a quark loop contracted onto a four-quark operator, plus the chromo-magnetic penguin operator $O_{8g}$—to two effective operators with coefficients $C^{\\rm eff}_{D,1}$ and $C^{\\rm eff}_{D,2}$. The loop function $G(m)$, decomposed into quark and mass terms, controls the size and imaginary part of the correction; its antisymmetric structure $F_V(p_g^2,\\tilde p_g^2)=-F_V(\\tilde p_g^2,p_g^2)$ means the correction contributes to $PP$ and longitudinal $VV$ final states but not to $PV$ at leading power. The NLO hard kernels $h_{\\rm nan}$ introduce new factorization formulas that are convolved with the $B$-meson and light-meson wave functions, including twist-2 and two-particle twist-3 light-cone distribution amplitudes; the endpoint-safe transverse-momentum-dependent framework of PQCD lets those convolutions be evaluated without parametric regularization.","core_discovery":"Within the PQCD factorization approach, the paper establishes that the quark-loop enhanced NLO contribution to pure-annihilation B decays is not a small perturbation of the branching ratios but is instead the dominant source of strong phases. The mechanism is a cancellation: the NLO amplitudes have large imaginary parts opposite in sign to the LO ones, so the modulus of the total amplitude and therefore the branching ratio barely change; for $\\bar B_s\\to\\pi^+\\pi^-$ and $\\bar B_s\\to\\rho_L^+\\rho_L^-$, $|A_{\\rm LO+NLO}|/|A_{\\rm LO}|\\approx 1.02$. The same imaginary parts convert into strong-phase differences between tree and penguin amplitudes ($\\sin\\Delta\\delta$ over $0.5$ for $\\bar B_s\\to\\pi\\pi$ and $\\bar B_s\\to\\rho\\rho$), and because CP asymmetries are proportional to $\\sin\\Delta\\delta\\,\\sin\\Delta\\phi$ (or to imaginary parts of amplitudes), they are strongly enhanced. Concretely, the direct asymmetry in $\\bar B_d\\to\\phi_L\\phi_L$ moves from $0$ to $-39.7\\%$, its mixing-induced asymmetry from $0$ to $27.8\\%$, and the $\\bar B_s\\to\\pi\\pi/\\rho\\rho/\\omega\\omega$ direct asymmetries go to about $-4\\%$ to $-6\\%$; $\\bar B_s\\to\\rho_L\\omega_L$ is the exception, remaining unchanged because its two-gluon production violates isospin conservation.","pith_inferences":["One consequence the paper leaves implicit: if the predicted large direct asymmetry in $\\bar B_d\\to\\phi_L\\phi_L$ survives a full NLO treatment, this mode could serve as a clean probe of the $b\\to d$ penguin phase, although its tiny branching ratio means only a high-luminosity experiment could test it.","The paper's assertion that the Fig. 3 triangle diagrams are negligible is not backed by a numerical estimate; computing those diagrams in the same framework would be the direct test of whether the quoted strong phases are stable or only approximate.","The same quark-loop mechanism should also shift the CP asymmetries of other penguin-dominated B decays that proceed through annihilation-type topologies, and the size of the shift could be predicted by applying the same effective-operator calculation to those channels.","The cancellation pattern—large imaginary NLO parts with opposite sign—suggests that the real part of the correction is what must be controlled to keep branching ratios stable, so improvements in the light-cone distribution amplitudes would mainly sharpen the CP predictions rather than the rates."],"forward_implications":["Branching-ratio predictions for the pure annihilation channels remain essentially at their leading-order PQCD values and stay compatible with current experimental bounds, so the new ingredient does not disturb the earlier successful rate predictions.","Direct CP asymmetries become measurable in several modes: about $-5\\%$ for $\\bar B_s\\to\\pi\\pi,\\rho\\rho,\\omega\\omega$, and about $-40\\%$ for $\\bar B_d\\to\\phi_L\\phi_L$, creating concrete targets for current and future B-physics experiments.","Mixing-induced asymmetries receive the larger shift because they are proportional to the imaginary part of the amplitude; for example, $\\bar B_s\\to\\pi\\pi$ $A_{\\rm CP}^{\\rm mix}$ moves from $+35.9\\%$ at LO to about $-4.2\\%$, a qualitative sign change.","The longitudinal polarization fraction in $B\\to VV$ modes stays near $1$, since the NLO correction only affects the longitudinal amplitude and the transverse amplitudes remain suppressed.","The $\\bar B_s\\to\\rho_L\\omega_L$ mode is predicted to keep its CP asymmetries unchanged, providing a control channel for the isospin-conservation argument behind the quark-loop contribution."],"supporting_citations":[{"why":"Supplies the earlier discovery that the charm-quark loop modifies the ratio of tree to penguin amplitudes and their strong phase; this paper extends it by including twist-3 LCDAs and more channels.","marker":"[47]"},{"why":"Provides the leading-order factorization formulas and light-meson LCDA inputs that the numerical evaluation here adopts.","marker":"[49]"},{"why":"Gives the LO PQCD amplitudes for the B to PP and B to VV pure annihilation decays on which the NLO corrections are built.","marker":"[10]"},{"why":"Defines the effective weak Hamiltonian, operator basis, and Wilson coefficients used for both LO and NLO amplitudes.","marker":"[48]"},{"why":"Establishes the PQCD transverse-momentum-dependent factorization framework that removes the endpoint singularities and makes the annihilation diagrams calculable.","marker":"[7]"}],"fun_headline_variants":["Quark-loop NLO turns off branching shifts, turns on CP violation","PQCD: NLO quark loops give B decays strong phases but no rate change","Cancellation keeps B rates, quark loops boost CP asymmetry to −40%","Pure annihilation B decays: NLO loops swap rates for CP asymmetry","Quark-loop strong phases drive B→φφ CP asymmetry to −40%"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The calculation assumes that the quark-loop and chromo-magnetic-penguin diagrams are the only important next-to-leading-order corrections, and that the triangle diagrams of Fig. 3 plus the QCD corrections to the leading-order diagrams really are negligible; the paper gives this as a qualitative argument, without a numerical estimate of the discarded terms.","fun_headline_variants_meta":{"raw":{"variants":["Quark-loop NLO turns off branching shifts, turns on CP violation","PQCD: NLO quark loops give B decays strong phases but no rate change","Cancellation keeps B rates, quark loops boost CP asymmetry to −40%","Pure annihilation B decays: NLO loops swap rates for CP asymmetry","Quark-loop strong phases drive B→φφ CP asymmetry to −40%"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000672,"raw_usage":{"total_tokens":3121,"prompt_tokens":1067,"completion_tokens":2054,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":683,"completion_tokens_details":{"reasoning_tokens":1954}},"tokens_in":683,"tokens_out":2054,"duration_ms":14266,"temperature":1.0,"reasoning_tokens":1954,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-16T11:35:15.392625+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Compute the triangle diagrams of Fig. 3 and the full QCD corrections to the Fig. 1 diagrams in the same PQCD framework: if their imaginary parts are comparable to or larger than the quark-loop contribution, the claimed dominance and the predicted strong phases collapse. On the experimental side, measuring $A_{\\rm CP}^{\\rm dir}(\\bar B_d\\to\\phi_L\\phi_L)$ near zero rather than the predicted about $-40\\%$ would falsify the central claim.","supporting_citations":[{"cited_title":"Enhanced Next-to-Leading-Order Corrections to Weak Annihilation $B$-Meson Decays","cited_arxiv_id":"2202.08073","evidence_quote":"Supplies the earlier discovery that the charm-quark loop modifies the ratio of tree to penguin amplitudes and their strong phase; this paper extends it by including twist-3 LCDAs and more channels."}],"review_version":1}