{"id":"d51d50a7-04e7-4ba4-995c-6f4b23fb6dce","arxiv_id":"1908.09398","paper_version":1,"verdict":"ACCEPT","confidence":"MODERATE","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":11,"one_line_summary":"A full heavy-quark expansion to order 1/m_c^2 for all B to D and D* form factors gives a consistent theory determination of |V_cb| = (40.3 ± 0.8) x 10^-3 and sharp predictions for R(D^(*)).","lead":"This paper computes all ten form factors that describe B meson decays to D or D* mesons within the heavy-quark expansion, including corrections of order 1/m_c^2. The results agree with experiment and yield a new value of the CKM element |V_cb| that is compatible with both exclusive and inclusive determinations.","discovery_kind":"extension","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The assumed power-counting truncation at O(ε^3) is untested, and the unusually low χ²/dof suggests the quoted |Vcb| uncertainty may be optimistic.","rationale":"The paper does what it claims: it assembles lattice, LCSR, QCD sum rule, and unitarity inputs and fits the full HQE form-factor set to O(1/m_c^2). The power-counting assumption is explicit, and the paper is transparent about its limitations. My primary concern is that the truncation and the covariance model are under-tested. The low χ² values (9.91 for 28 dof in the theory-only 3/2/1 fit; 30.00 for 55 dof in the full fit) suggest the quoted uncertainties may not be conservative; this matters because the headline |Vcb| error of ±0.8×10⁻³ is partly inherited from the theory covariance. The proposed expanded-model and inflated-covariance cross-checks would settle whether the conclusion is stable. I do not think this warrants rejection: the authors disclose the assumption, provide posterior samples for reanalysis, and the central SM predictions are not sensitive to the separately excluded f_T constraint. The f_T mismatch noted in Sec. III.A is a secondary caveat, slightly overstating the phrase 'full set of ten' for f_T, but it does not affect the |Vcb| extraction or the R(D(*)) predictions. Verdict remains ACCEPT unchanged, with the convergence and covariance robustness as the key items for future scrutiny.","tokens_in":16434,"tokens_out":15351,"duration_ms":146327,"concrete_test":"Re-run the nominal 3/2/1 analysis with (i) an expanded 4/3/2 model that adds one z-order to each class of Isgur-Wise functions, and (ii) a model that augments the LCSR covariance with an uncorrelated 30% systematic component. If the posterior-mean |Vcb| shifts by more than about 0.5×10⁻³, or if the 68% interval changes by more than about 30%, then the quoted (40.3 ± 0.8)×10⁻³ is not robust to the truncation and covariance assumptions. Additionally, compare the Bayesian log-evidence among the 2/1/0, 3/2/1, and 4/3/2 models; if the evidence does not saturate, the convergence claim is unsupported.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The load-bearing assumption is the power-counting ε_b ~ ε_c^2 ~ α_s/π ~ ε^2 with all mixed and higher-order terms taken as negligible (Sec. II.A). The nominal 3/2/1 model is only one z-order higher than the minimal 2/1/0 model, and neither model includes O(ε_b ε_c), O(α_s ε_c), or O(ε_c^3) contributions; the paper itself states this assumption should be revisited. The internal evidence does not prove the assumption: the theory-only 3/2/1 fit gives χ² = 9.91 for 28 dof, a p-value near 0.999, which is suspiciously good and suggests the input covariance (especially the LCSR errors at w ≥ 1.5) is overestimated. If those uncertainties are inflated, the posterior widths of the Isgur-Wise functions are not as constraining as they appear, and the 2% uncertainty on |Vcb| could be an artifact of the assumed covariance rather than a robust statement about convergence. Since the determination of all ten form factors and the subsequent |Vcb| extraction depend on the posterior width being meaningful, this is the weakest point in the central claim.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"This paper determines the full set of ten B->D^(*) semileptonic form factors within the Heavy-Quark Expansion (HQE) to O(alpha_s, 1/m_b, 1/m_c^2), using lattice QCD data, light-cone sum rules, QCD three-point sum rules, and unitarity bounds. The authors introduce two z-expansion models for the Isgur-Wise functions, denoted 2/1/0 and 3/2/1, and perform Bayesian fits to theory constraints with and without Belle experimental distributions. They report posterior predictions for R_D^(*) and tau polarization observables, and extract an exclusive |V_cb| = (40.3 +/- 0.8) x 10^-3 that is compatible with both the BGL-based exclusive value and the inclusive determination at the 1.2 sigma level. Posterior samples and the EOS-based analysis code are made publicly available.","tokens_in":16886,"tokens_out":5946,"duration_ms":59695,"significance":"If the result holds, this is a significant advance: it is the first determination of the complete set of B->D^(*) form factors at subsubleading power in the HQE, and it provides a theory-driven exclusive |V_cb| that is consistent with the inclusive value, thereby addressing the long-standing V_cb puzzle without assuming the absence of new physics in light-lepton modes. The public release of posterior samples and code is a clear strength of the paper. However, the robustness of the quoted precision depends on an untested power-counting truncation and on the statistical consistency of the input constraints, so the central claims should be re-examined in revision.","major_comments":[{"comment":"The central convergence claim is not supported to the precision claimed. The power counting epsilon_b ~ epsilon_c^2 ~ alpha_s/pi ~ epsilon^2 relegates all O(epsilon^3) and mixed terms to 'assumed to be negligible,' but no quantitative estimate of these terms is provided. The fitted 1/m_c^2 coefficients in Table II have very broad posteriors; for example, l_1'(1) = -5.78 with a 68% interval [-12.5,-0.16], and l_3(1) = 0.86 with [-8.29,5.17]. These intervals are compatible with a wide range of magnitudes and do not by themselves establish that the subsubleading coefficients are of O(1). The authors should estimate the size of neglected O(epsilon^3) contributions (for instance by evaluating the fitted IW functions at the edge of the recoil range) or explicitly add a truncation uncertainty to the quoted form factors and |V_cb|.","section":"Sec. II.A and Table II"},{"comment":"The goodness of fit for the theory-only fits is unusually good, indicating that the input uncertainties may be overestimated. The 3/2/1 theory-only fit gives chi^2 = 9.91 for 28 dof, corresponding to a p-value of about 0.999; the 2/1/0 theory-only fit has chi^2/dof = 22.87/38, also well below 1. Such low values suggest that the covariance of the inputs, most plausibly the LCSR constraints at w >= 1.5, is inflated or that correlations are not fully captured. Because the posterior widths of the Isgur-Wise functions directly feed into the quoted uncertainty of |V_cb| = (40.3 +/- 0.8) x 10^-3, the authors should address this mismatch, for example by reporting goodness-of-fit p-values, scaling the LCSR covariance, or demonstrating that the result is robust to a recalibration of those uncertainties.","section":"Sec. III.A and Table I"},{"comment":"The choice of the 3/2/1 model as the nominal model is not propagated as a systematic uncertainty. The combined |V_cb| changes from 40.7 +/- 1.0 in the 2/1/0 theory-only fit to 40.2 +/- 1.0 in the 3/2/1 theory-only fit, while the final quoted uncertainty is 0.8. The z-order truncation therefore shifts the central value by about half of the quoted error. The authors should include this model uncertainty, for instance by model averaging over the two z-orders or by adding the observed shift in quadrature to the statistical uncertainty.","section":"Sec. III.B and Table III"}],"minor_comments":[{"comment":"The legend of the P_D*(chi) panel lists 'fit 3/1/0', which appears to be a typo for 'fit 3/2/1'.","section":"Figure 2, P_D*(chi) panel"},{"comment":"The transition from the expansion in (w-1)^k to the expansion in z is described only verbally; a brief explicit definition of z(w) would improve clarity, especially because the maximum |z| used here is larger than in previous studies.","section":"Sec. II.A, eq. (10)"},{"comment":"The text contains a duplicated article: 'the the DFG Excellence Cluster' should read 'the DFG Excellence Cluster'.","section":"Acknowledgments"},{"comment":"The sentence 'Averaging the two exclusive determinations with the inclusive one' is slightly confusing because the exclusive determination is presented as a single combined result; rephrasing to refer to the D and D* channels individually would be clearer.","section":"Sec. III.B, after eq. (14)"},{"comment":"The caption states that uncertainty ranges are 'for illustrative purpose only' but these ranges are then used in the text to discuss the size of the 1/m_c^2 coefficients; a more precise statement about their interpretation and coverage would be helpful.","section":"Table II caption"}],"recommendation":"major_revision","confidential_remarks":null},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Two things to know. This is the first analysis to include all six subsubleading Isgur-Wise functions at O(1/m_c²) in a global fit of the ten B→D(*) form factors, and the first to extract |V_cb| at that order. It is also a model of reproducibility: two independent code implementations, public EOS code, and posterior samples shipped on Zenodo. That part is genuinely good.\n\nWhat it does well, concretely. The Gubernari–Kokulu–van Dyk LCSR input finally makes the full subsubleading set accessible, and the fit to lattice, QCD sum rules, LCSR, and unitarity bounds is clean. The central results — R(D), R(D*), F_L, and |V_cb| = (40.3 ± 0.8)×10⁻³ compatible with both BGL-exclusive and inclusive determinations at 1.2σ — hang together, and the predictions stay within ~1σ of earlier analyses while claiming smaller uncertainties. The authors are honest about the truncation: Section II.A states that higher-order and mixed terms are assumed negligible, and Section IV repeats that this assumption should be revisited as better data arrive.\n\nSoft spots, in proportion. The weakest point is not the power counting itself; it is the suspiciously good theory-only fit: χ² = 9.91 for 28 dof in the 3/2/1 model, a p-value near 0.999. That suggests the input covariance, most likely the LCSR errors at large w, is overestimated. If so, the posterior widths — and hence the ±0.8 on |V_cb| — are optimistic. This is an uncertainty-size problem, not a central-value problem, but it is exactly where a referee should push. Related, the convergence claim rests on coefficients like l1'(1) = −5.78 with a range [−12.5, −0.16]; calling those “of O(1)” is a generous reading of parameters that broad. And the choice of 3/2/1 over 2/1/0 as the nominal model is a systematic that is not propagated into the final numbers. The author overlap with the LCSR input and the EOS code is transparent and the outputs are public; I would not call it circular.\n\nBottom line: anyone working on semileptonic B decays, |V_cb|, or R(D^(*)) should read this, and it deserves a serious referee. I would send it to review, with a request to propagate the model-dependence systematic and take a hard look at the LCSR covariance.","headline":"A genuinely new, carefully executed HQE analysis of all ten B→D(*) form factors at O(1/m_c²) with a sensible |V_cb|; deserves a serious referee, though the quoted uncertainties are probably too small.","tokens_in":17400,"tokens_out":5663,"would_cite":true,"duration_ms":51625,"reading_group":"yes","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":[],"pacs":[],"model":"deepseek-v4-flash","headline":"Theory alone fixes the ten $\\bar B\\to D^{(*)}$ form factors and reconciles the exclusive and inclusive determinations of $|V_{cb}|$.","keywords":["heavy-quark expansion","B meson semileptonic decays","form factors","Isgur-Wise functions","|Vcb|","light-cone sum rules","lattice QCD","lepton flavour universality"],"falsifier":"A lattice QCD calculation of the $\\bar B\\to D^*$ form factors at nonzero recoil precise enough to resolve the subleading Isgur-Wise functions, or a measurement of the $\\cos\\theta_\\ell$ distribution in $\\bar B\\to D^*\\ell\\bar\\nu$ that disagrees with the posterior prediction, would test the expansion; alternatively, computing the $\\mathcal{O}(\\varepsilon_b\\varepsilon_c)$ mixed corrections and finding them to shift $|V_{cb}|$ by more than roughly $1\\times 10^{-3}$ would falsify the truncation.","tokens_in":16224,"feed_emoji":"⚛️","tokens_out":5204,"duration_ms":44921,"temperature":0.7,"pith_summary":"This paper establishes that the full set of ten form factors governing $\\bar{B}\\to D^{(*)}\\ell^-\\bar\\nu$ decays can be determined from theory inputs alone—lattice QCD, light-cone sum rules, QCD sum rules, and unitarity—within the Heavy-Quark Expansion to order $\\mathcal{O}(\\alpha_s, 1/m_b, 1/m_c^2)$. The coefficients of the $1/m_c^2$ terms come out of order one, which the authors read as evidence that the expansion converges at this order. With these form factors, the exclusive determination of the CKM element $|V_{cb}|$ from light-lepton decays is $(40.3\\pm 0.8)\\times 10^{-3}$, compatible with the inclusive determination at the $1.2\\sigma$ level. The same machinery yields precision predictions for the lepton-flavour-universality ratios $R_{D^{(*)}}$, the $\\tau$ polarizations $P_\\tau^{D^{(*)}}$, and the longitudinal polarization fraction $F_L$.","feed_headline":"Ten form factors from theory alone shrink the Vcb gap","feed_subtitle":"A heavy-quark expansion to 1/m_c^2 gives |Vcb| = (40.3 ± 0.8)×10^-3, now compatible with the inclusive value.","key_machinery":"The central object is the Heavy-Quark Expansion of the hadronic matrix elements under the power counting $\\varepsilon_b \\sim \\varepsilon_c^2 \\sim \\alpha_s/\\pi \\sim \\varepsilon^2$, where $\\varepsilon_Q = \\bar\\Lambda/2m_Q$. Within this scheme each form factor is expressed through ten independent Isgur-Wise functions—the leading $\\xi$, the subleading $\\chi_{2,3}$ and $\\eta$, and the six subsubleading $\\ell_1,\\dots,\\ell_6$—each expanded in the conformal variable $z$ around zero recoil. The argument is carried by a Bayesian fit that combines lattice pseudo-data, LCSR results for all form factors at $q^2\\lesssim 0$, three-point QCD sum rules for the subleading functions, and dispersive unitarity bounds, with the minimal viable z-expansion model 2/1/0 and the nominal model 3/2/1.","core_discovery":"The paper claims that the ten independent $\\bar{B}\\to D^{(*)}$ form factors are determined at physical recoil by the Heavy-Quark Expansion truncated at $\\mathcal{O}(\\alpha_s, 1/m_b, 1/m_c^2)$, with all ten Isgur-Wise functions at subsubleading power included for the first time. The key finding is that the subsubleading coefficients $\\hat{\\ell}_i$ are of $\\mathcal{O}(1)$, so the $1/m_c^2$ corrections are neither anomalously large nor accidentally small; the expansion is well behaved. Combining the fit with Belle kinematic distributions and HFLAV branching ratios gives $|V_{cb}| = (40.3\\pm 0.8)\\times 10^{-3}$, which agrees with both the BGL-based exclusive value and the inclusive determination at $1.2\\sigma$.","pith_inferences":["The same combined-analysis strategy—lattice plus LCSR plus unitarity in a z-expanded HQE—should transfer directly to other heavy-quark transitions such as $\\Lambda_b\\to\\Lambda_c \\ell\\bar\\nu$, where the subleading Isgur-Wise functions are less constrained.","The visible pull in $P(\\cos\\theta_\\ell)$ suggests that a future high-statistics measurement of the lepton angular distribution in $\\bar B\\to D^*\\ell\\bar\\nu$ is a sharper test of the HQE form-factor shapes than the recoil spectra alone.","If the neglected $\\mathcal{O}(\\varepsilon^3)$ and mixed $\\varepsilon_b\\varepsilon_c$ terms were included, the central value of $|V_{cb}|$ would shift at the level of the current uncertainty, so the quoted $1.2\\sigma$ compatibility with the inclusive value is itself contingent on the power counting."],"forward_implications":["The full set of form factors is available without assuming the absence of new physics in light-lepton modes, so $|V_{cb}|$ and new-physics Wilson coefficients can be extracted from the same data.","The $|V_{cb}|$ puzzle shrinks: the exclusive HQE value $(40.3\\pm 0.8)\\times 10^{-3}$ agrees with the inclusive determination at $1.2\\sigma$.","Predictions for $\\bar B\\to D^{(*)} \\tau\\bar\\nu$ observables sharpen: with experimental input the paper finds $R_D=0.297\\pm 0.003$ and $R_{D^*}=0.250\\pm 0.003$, with uncertainties well below current measurements.","The order-one size of the $1/m_c^2$ coefficients indicates that neglected $\\mathcal{O}(\\varepsilon^3)$ and mixed $\\varepsilon_b\\varepsilon_c$ terms are not visible at present precision.","Upcoming lattice results for $\\bar B\\to D^*$ form factors at nonzero recoil can be incorporated directly into the same framework to reduce uncertainties further."],"supporting_citations":[{"why":"LCSR calculation providing the first information on all ten form factors at and beyond maximal recoil, anchoring the subsubleading Isgur-Wise functions.","marker":"[9]"},{"why":"MILC lattice determination of the $\\bar B\\to D$ form factors $f_+$ and $f_0$ used as constraints.","marker":"[6]"},{"why":"HPQCD lattice determination of the same $\\bar B\\to D$ form factors, providing independent correlated pseudo-data.","marker":"[7]"},{"why":"HPQCD lattice calculation of the $\\bar B\\to D^*$ form factor $h_{A_1}$ at zero recoil.","marker":"[8]"},{"why":"Fermilab/MILC lattice determination of $h_{A_1}$ at zero recoil, combined with [8] in the FLAG average.","marker":"[25]"},{"why":"Establishes the HQE framework with leading and subleading Isgur-Wise functions including partial $1/m_c^2$ corrections that this work extends.","marker":"[12]"},{"why":"Prior HQE fit to $\\bar B\\to D^{(*)}$ constraints that this paper generalizes to the full set of subsubleading functions.","marker":"[13]"},{"why":"Boyd-Grinstein-Lebed dispersion-relation unitarity bounds, the basis for the BGL-style constraints used in the fit.","marker":"[10]"},{"why":"Caprini-Lellouch-Neubert formulation of unitarity bounds adapted to the HQE framework and the z expansion.","marker":"[24]"},{"why":"QCD three-point sum-rule determination of the subleading Isgur-Wise functions, providing the five QCDSR observations.","marker":"[27]"}],"fun_headline_variants":["Subleading 1/m_c^2 corrections reconcile Vcb values","Theory-only form factors close the exclusive-inclusive gap","HQE at 1/m_c^2: ten form factors, one consistent Vcb","O(1) subsubleading terms confirm HQE convergence","All ten B-to-D form factors from theory now match data"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The whole extraction rests on the power counting $\\varepsilon_b\\sim\\varepsilon_c^2\\sim\\alpha_s/\\pi\\sim\\varepsilon^2$ and on treating every higher-order or mixed term as negligible; if any of those omitted terms contributes at the current precision, the form factors and $|V_{cb}|$ shift by more than the quoted uncertainty.","fun_headline_variants_meta":{"raw":{"variants":["Subleading 1/m_c^2 corrections reconcile Vcb values","Theory-only form factors close the exclusive-inclusive gap","HQE at 1/m_c^2: ten form factors, one consistent Vcb","O(1) subsubleading terms confirm HQE convergence","All ten B-to-D form factors from theory now match data"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000227,"raw_usage":{"total_tokens":1512,"prompt_tokens":1028,"completion_tokens":484,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":644,"completion_tokens_details":{"reasoning_tokens":393}},"tokens_in":644,"tokens_out":484,"duration_ms":5397,"temperature":1.0,"reasoning_tokens":393,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-14T11:12:50.686319+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"A lattice QCD calculation of the $\\bar B\\to D^*$ form factors at nonzero recoil precise enough to resolve the subleading Isgur-Wise functions, or a measurement of the $\\cos\\theta_\\ell$ distribution in $\\bar B\\to D^*\\ell\\bar\\nu$ that disagrees with the posterior prediction, would test the expansion; alternatively, computing the $\\mathcal{O}(\\varepsilon_b\\varepsilon_c)$ mixed corrections and finding them to shift $|V_{cb}|$ by more than roughly $1\\times 10^{-3}$ would falsify the truncation.","supporting_citations":[],"review_version":1}