{"id":"0be15d29-2968-4f9c-a94a-ba0371f9f7ec","arxiv_id":"2510.03739","paper_version":2,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":4.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":3,"one_line_summary":"Using the covariant confined quark model, the paper predicts branching fraction, F_L, A_FB and optimized angular observables for B_s → φ ℓ+ℓ− and finds broad agreement with LHCb data.","lead":"A quark-model paper computes the rate and angular structure of the rare decay B_s → φ ℓ+ℓ− and compares it with LHCb measurements. It is a Standard-Model consistency check in a channel where a >3σ deviation was once reported.","discovery_kind":"extension","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Table VI, the sole LHCb-comparison table, contains misaligned rows that currently undermine the claimed agreement.","rationale":"I read the paper in good faith as an application of the CCQM to B_s→φℓ+ℓ− with new binned angular observables and NNLL corrections. The most load-bearing condition for the central claim is the integrity of Table VI, because the abstract's 'agreement with the latest experimental data' is supported almost entirely by that table. The quoted entries include internally inconsistent values that cannot all be correct: AFB[1,6] cannot equal FL[1,6], A5[0.1,0.98] cannot be two orders of magnitude larger than A5[0.1,2], and A8[1,6] cannot equal S4[1,6]. This is not an attack on the physics framework or on the authors; it is a concrete, checkable defect in the evidence. The reader's weakest_assumption focused on inherited form factors and charm-loop modeling. Those are legitimate secondary concerns, but the table corruption is more immediate: even if every model input is correct, the paper as printed does not yet demonstrate agreement. The issue is fixable by regenerating the table, so the conditional verdict remains appropriate rather than a rejection. I would not change the reader's verdict, hence 'UNCHANGED'.","tokens_in":15439,"tokens_out":10499,"duration_ms":176685,"concrete_test":"Recompute Table VI from Eqs. (9)–(19) with the Table I form factors and the stated Wilson coefficients. Specifically evaluate A5 in [0.1,0.98] and AFB in [1,6]/[15,18.9], and cross-check the LHCb columns against refs. [1,2]. If A5 comes out near 0.003 instead of 0.268, or AFB comes out near 0.03 instead of 0.69, the table is misaligned. Then regenerate all rows and re-evaluate the claimed agreement.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central claim is the LHCb agreement in Table VI, but that table has demonstrable row/column misalignments. Examples: (i) AFB block row [1,6] lists CCQM = 0.69±0.14 and PQCD = 0.777 — identical to the FL values for the same bin, while their own [1.1,6] row has AFB = 0.034±0.006; (ii) A5 [0.1,0.98] = 0.268±0.054 while adjacent [0.1,2] = 0.0031±0.0006, an impossible jump for a smooth angular observable; (iii) A8 [1,6] = 0.17±0.03 equals the S4 [1,6] entry, while A8 should be tiny; (iv) S7 rows are missing CCQM entries and contain PQCD values of order 0.45, incompatible with S7's expected scale. Because the paper's abstract and conclusion rest on this table, the agreement claim is not substantiated as printed. The issue is a correctable presentation error, but it must be fixed before the numerical comparison can be evaluated.","agreement_with_reader":"partial"},"referee_report":{"model":"deepseek-v4-flash","summary":"This paper presents a calculation of the B_s→φ transition form factors in the covariant confined quark model (CCQM), and uses them to predict branching fractions and angular observables (A_FB, F_L, A_i, S_i) for B_s→φℓ^+ℓ^- in various q² bins. The results are compared with LHCb data from 2015 and 2021 and with PQCD predictions from ref. [19]. The central claim is that the CCQM predictions agree with the latest LHCb data, particularly in specific high-q² bins, and that NNLL two-loop corrections to C7_eff and C9_eff are essential to this agreement. The manuscript includes tables of binned predictions, figures for differential distributions, and a comparison of integrated observables.","tokens_in":15754,"tokens_out":6587,"duration_ms":52091,"significance":"If the numerical results are correct, this would be a useful independent CCQM benchmark for b→sℓℓ transitions and a cross-check of the LHCb angular measurements. The paper's use of externally computed Wilson coefficients and published two-loop corrections is a strength, as is the direct comparison with two LHCb datasets. However, the main comparison table is substantially misaligned and the text overstates the level of agreement. The quantitative conclusions cannot be assessed until the table is corrected and the internal inconsistencies are resolved.","major_comments":[{"comment":"Table VI, the sole LHCb-comparison table, contains multiple row/column misalignments that invalidate the printed agreement claim. In the AFB block, row [1,6] lists CCQM = 0.69 ± 0.14 and PQCD = 0.777, identical to the FL values for the same bin, whereas the adjacent [1.1,6] AFB row is 0.034 ± 0.006. In the same AFB block, row [15,18.9] lists 0.394 ± 0.003, again an FL value. A5 [0.1,0.98] = 0.268 ± 0.054 while A5 [0.1,2] = 0.0031 ± 0.0006, an implausible jump for a smooth observable. A8 [1,6] = 0.17 ± 0.03 equals the S4 entry. Many S7 rows lack CCQM entries and instead list PQCD values around 0.45–0.99, inconsistent with the stated magnitude of S7. These errors must be corrected before the agreement claim can be evaluated.","section":"Table VI"},{"comment":"The concluding sentence in Section V states that the integrated observables are in 'complete agreement' with PQCD [19]. This is contradicted by Table V: for the τ mode, <F_L> is 0.090 ± 0.02 (CCQM) versus 0.396+0.002/−0.003 (PQCD), a factor-of-4 difference. The μ-mode values are close, but the τ-mode F_L is not in agreement. This overstatement should be removed and the discrepancy discussed.","section":"Section V"},{"comment":"The paper states that the [5,8] and [6,8] GeV² bins are omitted from NNLL corrections and therefore use one-loop Wilson coefficients, yet these bins are included in Table VI without any additional systematic uncertainty. Since the text credits NNLL corrections for the agreement with LHCb, the one-loop bins are not on the same footing as the rest of the table. The authors should either assign an uncertainty for the missing two-loop piece in these bins or explicitly exclude them from the agreement claim.","section":"Section IV, Eq. (20)"},{"comment":"The manuscript gives two conflicting validity ranges for the low-q² NNLL corrections. Equation (20) states 1.1 ≤ q² ≤ 5.5 GeV², while the following paragraph says the low-q² two-loop results are reliable for 0.1 ≤ q² ≤ 6 GeV². Several bins in Table VI start at 0.1 or 0.98 GeV². Because the paper argues that NNLL corrections are essential to the agreement, the reader must know which bins actually received these corrections. Please reconcile the ranges and state the treatment of each bin.","section":"Eq. (20)"}],"minor_comments":[{"comment":"The figure labels contain LaTeX artifacts such as '/LParen1GeV2/RParen1' and '/MiΝus'. The sentence before Fig. 1, 'The behavior of the differential branching B(Bs→φνν) is shown in Figs. 4', is incomplete.","section":"Figures 1–6"},{"comment":"Table I gives A0(0)=0.40, while Table II lists the CCQM A0(0)=0.28±0.03. The relation between these quantities (via the BSW form-factor convention from ref. [5]) should be stated explicitly in the caption of Table II, otherwise the reader cannot reconcile the two tables.","section":"Tables I–II"},{"comment":"The text says 'the optimized observable P1 remains small across a large range of values' and that A0(0)−V(0)=0.09 leads to a 'truly small value of P1'. However, Table V reports <P1> = −0.52 ± 0.1 for the muon mode. This is not small in the conventional normalization of P1 and should be clarified.","section":"Section IV"},{"comment":"The branching ratios in Table IV are quoted without explicit reference to the q² integration range. For Bs→φµ+µ−, the total rate should be specified as the full q² range; otherwise the comparison with LHCb and other models is ambiguous.","section":"Table IV"}],"recommendation":"major_revision","confidential_remarks":"The manuscript seems hastily assembled: Table VI has extensive row misalignments and several internal inconsistencies. I recommend that the authors verify all entries in Tables V and VI numerically before resubmission, because the central LHCb-agreement claim currently rests on a table that cannot be trusted as printed."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"You should know: this is a straightforward CCQM application to B_s → φ ℓ+ℓ−, reusing the form factors from the author's earlier paper. The genuinely new part is the binned angular observables with NNLL corrections and the comparison with LHCb 2021. But the paper's central claim—agreement with LHCb—is not actually substantiated as printed, because the key table (Table VI) contains misaligned rows and repeated entries that look like copy-paste errors.\n\nWhat the paper does well: it provides an independent hadronic-model cross-check of the Standard Model in a decay that has had LHCb tension, and it does so without fitting to the data. The effective-Hamiltonian machinery is standard, the Wilson coefficients come from external sources, and the two-loop correction regions are stated explicitly. The inclusion of the tau mode and the comparison with PQCD are useful, even if not novel. It is honest that the [5,8] and [6,8] bins use one-loop corrections with no assigned systematic.\n\nThe soft spots are real and load-bearing. The stress-test note is correct: Table VI has rows that simply repeat the wrong observable. For example, the AFB block row for [1,6] lists 0.69±0.14 and 0.777, which are the FL values for that bin, while their own [1.1,6] row gives a completely different AFB. A5 [0.1,0.98] jumps from 0.268 to 0.0031 in the adjacent [0.1,2] bin, which is implausible for a smooth angular observable. A8 [1,6] equals the S4 entry. S7 has no CCQM entries in several rows and PQCD values of order 0.45, which is incompatible with S7's expected scale. These are exactly the rows the abstract and conclusion lean on, so the 'agreement with LHCb' claim is unsupported as printed. Also, Table V says the tau-mode <F_L> is 0.090±0.02 in CCQM versus 0.396 in PQCD, while the text claims 'complete agreement.' That is not agreement, and it needs to be fixed or explained.\n\nNone of this looks like a broken central derivation—the form-factor parametrization and effective Hamiltonian are fine—but the presentation errors are severe enough that no quantitative conclusion can be drawn from this version. The reproducibility is also low: no code, no model-parameter table, though the parameters are in the cited previous work.\n\nWho this is for: someone working on B → V ℓ+ℓ− phenomenology who wants a CCQM cross-check of the SM. It deserves a serious referee because the physics is relevant and the errors are fixable, but only after the tables are corrected. I'd send it back for major revision, not desk-reject.","headline":"A useful CCQM cross-check undermined by a corrupted comparison table and an overclaimed agreement sentence.","tokens_in":16281,"tokens_out":1933,"would_cite":false,"duration_ms":16756,"reading_group":"maybe","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":[],"pacs":["12.39.Ki","13.30.Ce","14.40.Nd"],"model":"deepseek-v4-flash","headline":"The covariant confined quark model reproduces LHCb angular data for B_s→φμ+μ− once two-loop corrections to the Wilson coefficients are included.","keywords":["B_s decay","rare B decays","angular observables","covariant confined quark model","effective Hamiltonian","Wilson coefficients","LHCb","flavor physics"],"falsifier":"A determination of B_s→φμ+μ− in the q² bins [5,8] and [6,8] GeV² with the same two-loop accuracy (or an experimental measurement with uncertainty below ~5%) would test the one-loop CCQM predictions directly; if the measured branching fraction in these bins deviates by more than the model's ~20% uncertainty while the two-loop-valid bins agree, the assumed treatment of the charm region would be falsified. Alternatively, a lattice QCD calculation of the B_s→φ form factors at q² = 0 that yields A0(0)=0.40 and V(0)=0.31 within errors would support the CCQM input.","tokens_in":15276,"feed_emoji":"⚛️","tokens_out":4049,"duration_ms":29246,"temperature":0.7,"pith_summary":"This paper argues that the rare decay B_s→φℓ+ℓ− can be described within the Standard Model by the covariant confined quark model (CCQM), a relativistic quark model, provided the effective Wilson coefficients C7 and C9 are upgraded with next-to-next-to-leading-logarithmic (NNLL) two-loop corrections. Using form factors computed in the CCQM and the standard effective Hamiltonian, the paper computes the branching fraction, forward-backward asymmetry, longitudinal polarization, and optimized angular observables in q² bins, and compares them to LHCb measurements. The central result is that the CCQM predictions for the branching fraction and F_L agree with the 2021 LHCb data in the bins where the two-loop corrections are valid, indicating that no new physics is required for these observables. If correct, this strengthens the case that a simple quark-model approach can capture the dynamics of b→sℓℓ transitions.","feed_headline":"Quark model matches LHCb B_s→φμμ data with 2-loop terms","feed_subtitle":"Adding NNLL corrections to the Wilson coefficients brings predicted branching ratios and polarizations into agreement with 2021 measurements","key_machinery":"The central object is the set of dimensionless form factors A0, A+, A−, V, a0, a+, g for the B_s→φ transition, parametrized by a dipole form F(q²)=F(0)/(1−a s + b s²) with s=q²/m_{Bs}². These feed the helicity amplitudes that define the observables. The second key ingredient is the effective Hamiltonian with Wilson coefficients, where the long-distance charm-loop effects and two-loop QCD corrections are absorbed into C7_eff and C9_eff; the two-loop corrections are included only in the q² ranges where their expansions are valid, 1.1–5.5 and 8.8–22 GeV². The combination of these ingredients produces predictions that match LHCb data.","core_discovery":"The paper computes the B_s→φ transition form factors within the CCQM and uses them together with the SM effective Hamiltonian to predict the differential branching ratio and angular observables for B_s→φℓ+ℓ−. It shows that after including the NNLL two-loop corrections to C7_eff and C9_eff, the predicted branching fraction B(B_s→φμ+μ−) and the longitudinal polarization F_L fall within the 2021 LHCb measurements in the bins [11,12.5], [15,17], [17,18.9], and [15,18.9] GeV², improving on the agreement with the earlier 2015 data. The paper concludes that the CCQM provides a consistent Standard Model description of this decay.","pith_inferences":["The paper's reliance on form factors from a previous calculation without recomputation means the numerical agreement is conditional on those model parameters; a public release or independent recalculation of the CCQM parameters would allow direct testing.","Because the two-loop corrections are valid only in specific q² ranges, the 'agreement' claim is strongest in those bins; future high-precision data in the [5,8] and [6,8] bins would sharply discriminate the CCQM's one-loop treatment from other models.","The near-equality A0(0)≈V(0) noted in the heavy-quark limit is not exact in the CCQM, which explains the smallness of P1; testing whether P1 remains small at higher q² in other models could probe the model's form-factor structure.","If the charm-loop Breit-Wigner ansatz is inadequate, the agreement could be accidental; lattice QCD form factors combined with the same effective Hamiltonian would provide a cross-check."],"forward_implications":["If the CCQM description holds, the measured B_s→φμ+μ− observables are consistent with the Standard Model, and deviations seen in earlier bins may be attributed to missing higher-order corrections rather than new physics.","The inclusion of NNLL corrections is necessary for agreement; one-loop predictions are insufficient, especially at low q².","The optimized observable S7, which vanishes at leading order, becomes non-zero with the two-loop corrections, providing a testable SM prediction.","For the bins [5,8] and [6,8] GeV², where two-loop corrections are not valid, the one-loop CCQM predictions are the best available and can be improved when full NNLO results for the charm region become available.","The same framework yields predictions for the τ+τ− mode and B_s→φνν̄, which can be checked by future experiments."],"fun_headline_variants":["Quark model plus NNLL matches LHCb B_s→φμμ data","2-loop Wilson terms align B_s→φ decay with LHCb","CCQM predictions fit LHCb bins after NNLL corrections","B_s→φℓ+ℓ− branching ratios now match 2021 LHCb","NNLL corrections bring B_s→φ observables in line"],"cache_read_input_tokens":2304,"weakest_assumption_plain":"The numerical predictions inherit the CCQM form factors and model parameters from an earlier paper without recomputation, so the agreement with LHCb data is conditional on those priors being correct; additionally, the long-distance charm-loop effects are modeled by a Breit-Wigner ansatz rather than computed from first principles.","fun_headline_variants_meta":{"raw":{"variants":["Quark model plus NNLL matches LHCb B_s→φμμ data","2-loop Wilson terms align B_s→φ decay with LHCb","CCQM predictions fit LHCb bins after NNLL corrections","B_s→φℓ+ℓ− branching ratios now match 2021 LHCb","NNLL corrections bring B_s→φ observables in line"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000147,"raw_usage":{"total_tokens":981,"prompt_tokens":658,"completion_tokens":323,"prompt_tokens_details":{"cached_tokens":256},"prompt_cache_hit_tokens":256,"prompt_cache_miss_tokens":402,"completion_tokens_details":{"reasoning_tokens":224}},"tokens_in":402,"tokens_out":323,"duration_ms":36044,"temperature":1.0,"reasoning_tokens":224,"cache_read_input_tokens":256,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-04T11:36:21.699082+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"A determination of B_s→φμ+μ− in the q² bins [5,8] and [6,8] GeV² with the same two-loop accuracy (or an experimental measurement with uncertainty below ~5%) would test the one-loop CCQM predictions directly; if the measured branching fraction in these bins deviates by more than the model's ~20% uncertainty while the two-loop-valid bins agree, the assumed treatment of the charm region would be falsified. Alternatively, a lattice QCD calculation of the B_s→φ form factors at q² = 0 that yields A0(0)=0.40 and V(0)=0.31 within errors would support the CCQM input.","supporting_citations":[],"review_version":1}