{"id":"5f0b6489-464c-44ff-8fb1-84e1b40be04c","arxiv_id":"2512.18569","paper_version":2,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":5.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":3,"one_line_summary":"A covariant quark-model calculation finds that the fully-charm tetraquark candidate X(6900), decaying to two J/ψ mesons, prefers an axial-vector–axial-vector [cc][c̄c̄] internal coupling over a vector–vector one.","lead":"A model calculation of the decay of the tetraquark candidate X(6900) into two J/ψ mesons favors an axial-vector–axial-vector arrangement of its two charm-quark pairs over a vector–vector one, matching the measured width. It is a step toward pinning down the internal structure of a fully-charm exotic hadron.","discovery_kind":"new_application","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The A-A-over-V-V conclusion depends on an unproved cancellation: Eqs. (41)/(44) assert M_b=M_c=0 with no derivation; if the crossed diagrams contribute, both the width and the hierarchy can change.","rationale":"The paper is a standard covariant-quark-model calculation: g_X is fixed by the compositeness condition (Eq. 21), g_J/ψ similarly (Eq. 34), and the input parameters are explicit. The Gaussian vertex is admittedly a model input; the Λ scan is some evidence of robustness, but it does not cover functional-form variation. I do not object to the model per se; the issue is that the amplitude used for both structures is only the direct diagram, and the claimed zero of the other two diagrams is a single unproved sentence. This is precisely the kind of discrete structural assumption that a referee should ask to be demonstrated, because a nonzero M_b+M_c would change the interference pattern and the hierarchy. The abstract's separate 2η_c prediction (66–88 keV) is also unsupported in the body, but it is not the central claim; it should be derived or removed. Given the reader's CONDITIONAL verdict already asks for this check, my stress-test does not move the verdict; it sharpens which check matters most.","tokens_in":12304,"tokens_out":9896,"duration_ms":106972,"concrete_test":"Derive the Feynman rules from Eq. (5) for the A-A and V-V currents, including color indices and the (Γ1↔Γ2) symmetrization, and numerically evaluate M_b and M_c with the same Gaussian vertices, propagators, and masses used for Fig. 3 (m_X=6.905, m_J/ψ=3.0969, m_c=1.67 GeV; Λ_X=6.8–7.2, Λ_J/ψ=2.8–3.2). If |M_b+M_c| is not zero to numerical precision (≲1% of |M_a|), Eqs. (41)/(44) fail and the widths in Fig. 3 must be recomputed including all three diagrams.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central claim is the comparison in Fig. 3: A-A gives Γ≈40–90 MeV, V-V gives Γ≈10–30 MeV, so A-A is 'consistent with experiments.' This comparison is built on the assertion in Eqs. (41) and (44) that the two crossed diagrams are exactly zero, M_b=M_c=0. The paper's only justification is a sentence in Sec. IV about cancellation 'such as c_b(x2), \\bar c_a(x3) and c_a(x1), \\bar c_b(x4) in Eq. (5)'; no color factor, trace identity, or integral argument is supplied. If these diagrams are nonzero, the amplitude is not the single box integral in Eqs. (40)/(43), and the A-A/V-V difference — which comes from the sign of the interference term in Eqs. (47)/(48) — can shift, not just the overall normalization. The paper itself flags the related model dependence of φ_X(k²)=exp(k²/Λ_X²) ('hard to calculate', Sec. II A) and concedes 'Unless Λ_X is very small and/or Λ_J/ψ is very large, the V-V coupling cannot [be] consistent' (Sec. IV), but the discrete cancellation is the least secure link: it is both load-bearing and unverified.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper calculates the decay width of the presumed tetraquark state X(6900) into J/psi J/psi in a covariant quark model with a diquark-antidiquark [cc][cbar cbar] current. Two possible Dirac structures are compared: vector-vector (V-V) and axial-vector-axial-vector (A-A) coupling. The coupling constants g_X and g_J/psi are fixed through compositeness conditions, not fitted to the observed width. The paper finds that the A-A coupling yields a width of roughly 40–90 MeV, which it claims is consistent with the measured X(6900)->2J/psi width, while the V-V coupling yields only about 10–30 MeV and is too small. The authors conclude that X(6900) is more likely to have the A-A internal structure. An additional prediction for X(6900)->2 eta_c, 66–88 keV, is stated in the abstract.","tokens_in":12670,"tokens_out":4244,"duration_ms":52988,"significance":"If the result holds, it provides a concrete, model-based discrimination between two plausible four-quark Dirac currents for the J^PC=2++ X(6900), going beyond mass-only calculations. The compositeness-condition treatment of g_X and g_J/psi is a strength: it avoids fitting the decay width, and the parameter scan over Lambda_X and Lambda_J/psi is transparent. The analytic expressions for the vertex functions and widths are given in sufficient detail that the calculation could, in principle, be reproduced. However, the central claim rests on two fragile pillars: the asserted cancellation of the crossed Feynman diagrams, and the absolute normalization set by an ad hoc Gaussian vertex ansatz. The significance is therefore conditional on these two points being properly justified.","major_comments":[{"comment":"The vanishing of the crossed diagrams, M_b = M_c = 0, is asserted without a derivation. The only explanation in Sec. IV is a garbled sentence about quarks cancelling 'such as c_b(x2), bar c_a(x3) and c_a(x1), bar c_b(x4)'. No color factor, trace identity, or integral argument is supplied. This cancellation is load-bearing: it reduces the full amplitude to the single box diagram of Eqs. (40)/(43), and the difference between the A-A and V-V widths—the sign of the interference term in Eqs. (47)/(48)—is entirely carried by that box. If the crossed diagrams are nonzero, the width formulas and the A-A/V-V hierarchy can change. Please provide an explicit demonstration, or a numerical evaluation showing that the crossed diagrams are negligible.","section":"Sec. III, Eqs. (41), (44); Sec. IV"},{"comment":"The absolute decay width—and hence the statement that A-A is 'consistent with the experiments'—is controlled by the assumed Gaussian vertex functions, with Lambda_X chosen by hand as 6.8–7.2 GeV and Lambda_J/psi scanned over 2.8–3.2 GeV. These functional forms and scales are taken from the model literature, not derived from QCD. The paper concedes that V-V could become consistent if Lambda_X is very small and/or Lambda_J/psi is very large, but the same caveat applies to the A-A conclusion: a different vertex shape or a Lambda_X outside the quoted band could change the normalization significantly. The parameter scan alone does not bound this model dependence. Please either quantify the sensitivity to the vertex ansatz or temper the claim that the A-A result is uniquely consistent with experiment.","section":"Eq. (16), Eq. (31), Sec. IV, Fig. 3"},{"comment":"The abstract states an additional prediction, Gamma(X(6900)->2 eta_c) = 66–88 keV, but nowhere in the manuscript is this decay calculated, nor are any formulas, inputs, or numerical results for eta_c given. This is not a presentation detail: it is a claim of a quantitative result that cannot be checked. Either add the calculation and results, or remove the eta_c prediction from the abstract.","section":"Abstract"}],"minor_comments":[{"comment":"Typo: 'its mass is inconsistent with the the above experimental results' should read 'with the above experimental results'.","section":"Sec. I, paragraph 2"},{"comment":"The Gaussian form is written as phi_X(k^2)=exp(k^2/Lambda_X^2). Since the text says it should decrease rapidly in the ultraviolet in Euclidean space, please clarify the sign convention: as written, with a Euclidean k^2, the exponential grows. Presumably k^2 is the Minkowski argument or the Euclidean k^2 enters with a minus sign.","section":"Sec. II A, Eq. (16)"},{"comment":"The sentence 'Figures 1b and 1c is zero' is grammatically unclear; also the explanation of the cancellation is hard to follow. Rewrite to state precisely which quark lines are contracted and why the color/trace factors cancel.","section":"Sec. IV, last paragraph"},{"comment":"The statement R1=2N1, R2=2N2 'but with different g_X' is confusing: the definitions of N1,N2 already include g_X. Please clarify the relation between the scalar integrals used in Eqs. (39) and (42) and the explicit forms in the appendix.","section":"Appendix A, Eq. (A12)"},{"comment":"The vertical axis label is missing; the caption says 'Decay width' but the quantity plotted (presumably Gamma in MeV) should be stated on the axis or in the caption.","section":"Fig. 3"}],"recommendation":"major_revision","confidential_remarks":"The paper is a straightforward application of the covariant quark model and is not circular: the width is not fed back into the model. The main technical risk is the unproved cancellation of the crossed diagrams, which is the single most load-bearing step of the calculation. The abstract's eta_c prediction is also absent from the body. Both need to be fixed before I can recommend acceptance. The modeling of the vertex function is a known limitation of this approach; it should be framed as such, not as a decisive experimental discrimination."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Quick take on arXiv:2512.18569. It is exactly what it says: a covariant quark model calculation of Gamma(X(6900)->2J/psi), comparing axial-vector–axial-vector and vector–vector internal couplings. The new content is modest but real: this is the first CQM application to X(6900), and the comparative A-A vs V-V calculation is not in the cited literature. The machinery is standard (nonlocal vertices, compositeness condition), and the paper does a decent job: the width for A-A sits at 40–90 MeV, comfortably overlapping the measured values; V-V sits at 10–30 MeV, too small; and the hierarchy is stable over the Lambda_X and Lambda_J/psi scan. Given that the only free inputs are the size parameters, that stability is a legitimate argument and not tuning.\n\nThe soft spots are real. The strongest is Eqs. (41) and (44), which set the two crossed diagrams M_b and M_c to zero. The only justification is a garbled sentence in Sec. IV about cancellations between quark lines, with no color, trace, or integral argument. This is load-bearing: the whole difference between A-A and V-V in Eqs. (47) and (48) comes from the sign of the interference term. If the crossed diagrams do not vanish, the hierarchy could shift, not just the normalization. The authors need to show the cancellation or verify it numerically. I would not insist on code, but I would insist on a derivation.\n\nA second issue: the abstract predicts Gamma(X->2 eta_c) = 66–88 keV, but I could not find any derivation of that number in the body. It is not in Section IV or in the appendix. Either the derivation is missing or the prediction belongs in the abstract only, which is not acceptable. Remove it or derive it.\n\nThe model dependence of the Gaussian vertex is honestly acknowledged: they admit phi_X is “hard to calculate” and that V-V could be made consistent if Lambda_X is very small or Lambda_J/psi very large. That is a fair concession, and the scan suggests typical values avoid that corner. The citation pattern is fine: experimental papers, the standard CQM references, and relevant spectroscopy literature are all there. I found no self-citation padding.\n\nWho is this for? Hadron spectroscopists working on fully heavy tetraquarks, particularly those interested in distinguishing coupling structures through decay widths. It deserves a serious referee. It is a model calculation, not a theorem, but the claim is clear, the comparison is meaningful, and the flaws are fixable. If the cancellation is confirmed and the abstract is fixed, this is a publishable, useful paper.\n\nRecommendation: send it to a capable referee, with a request to check the M_b=M_c=0 step and the scalar integrals.","headline":"Reasonable CQM calculation that prefers the A-A coupling for X(6900) based on the decay width; the preference is plausible but rests on an unproved cancellation of two crossed diagrams, and the abstract's 2-eta_c prediction is absent from the body.","tokens_in":13199,"tokens_out":3329,"would_cite":true,"duration_ms":34250,"reading_group":"maybe","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":[],"pacs":[],"model":"deepseek-v4-flash","headline":"The X(6900) tetraquark's internal coupling is likely axial-vector–axial-vector, not vector–vector.","keywords":["X(6900)","fully charmed tetraquark","diquark-antidiquark","covariant quark model","decay width","J/psi pair","axial-vector coupling","tetraquark structure"],"falsifier":"Compute the crossed diagram contributions M_b + M_c explicitly (or replace the Gaussian vertex with a dipole form) and check whether the A-A width stays above the ~40 MeV lower edge of the data; if the cancellation fails or the shape change pulls the width below the measured values, the axial-vector–axial-vector conclusion would be weakened.","tokens_in":12136,"feed_emoji":"⚛️","tokens_out":2575,"duration_ms":27967,"temperature":0.7,"pith_summary":"This paper calculates the decay width of the fully charmed tetraquark candidate X(6900) into a pair of J/ψ mesons, assuming the state has a diquark–antidiquark [cc][c̄c̄] structure with spin-parity 2⁺⁺. Two possible Dirac couplings between the four constituent quarks are tested: vector–vector and axial-vector–axial-vector. The computed width for the axial-vector–axial-vector coupling lands in the 40–90 MeV range, matching the values observed by LHCb, ATLAS, and CMS, while the vector–vector coupling gives only 10–30 MeV, too small to explain the data. The paper therefore concludes that X(6900) is more likely built from axial-vector diquarks, and it predicts a decay width of 66–88 keV for the related X(6900)→2ηc channel.","feed_headline":"Axial-vector coupling wins for tetraquark X(6900)","feed_subtitle":"Model decay widths to 2J/ψ match measured 80–190 MeV only for the axial-vector–axial-vector structure.","key_machinery":"The central object is the nonlocal four-quark interpolating current for a diquark–antidiquark tetraquark, J^μν_X = [c_a^T C Γ₁ c_b][c̄_a Γ₂ C c̄_b^T] + (Γ₁↔Γ₂), with Γ₁,₂ chosen either as γ^μ⊗γ^ν (axial-vector–axial-vector) or γ^μγ⁵⊗γ^νγ⁵ (vector–vector). The vertex function is a Gaussian with scale Λ_X ≈ 6.8–7.2 GeV, and the coupling constants g_X and g_{J/ψ} are fixed by the compositeness condition. The decay amplitude is built from three Feynman diagrams, but two of them (the crossed quark-loop diagrams) are found to cancel exactly, leaving a single diagram whose scalar integrals N₁, N₂ (or R₁, R₂) determine the width.","core_discovery":"In a covariant quark model with nonlocal four-quark currents, the decay X(6900)→2J/ψ is computed for two spin-2 tetraquark currents: the axial-vector–axial-vector (A-A) current Γ₁⊗Γ₂ = γ^μ⊗γ^ν and the vector–vector (V-V) current Γ₁⊗Γ₂ = γ^μγ⁵⊗γ^νγ⁵. The coupling constants are fixed by the compositeness condition, and the vertex is a Gaussian in the relative momenta. The A-A coupling yields a partial width that is clearly larger than the V-V result and consistent with the measured X(6900) widths across the three experiments, whereas the V-V coupling is too small unless the model parameters are pushed to extreme values. From this the paper argues that the internal structure of X(6900) is more","pith_inferences":["The absolute-width comparison hinges on the Gaussian vertex ansatz: a different functional form (e.g., a dipole) could rescale both A-A and V-V widths and potentially shift the verdict, though the A-A-over-V-V hierarchy might survive if the cancellation of crossed diagrams persists.","The asserted exact cancellation of the two crossed diagrams (M_b = M_c = 0) is a strong simplification; verifying it explicitly with momentum-dependent vertices would test whether the single-diagram dominance holds beyond the present setup.","The predicted 2ηc width could be measured at the LHC or a future collider; if it lands outside 66–88 keV, the common coupling constants inferred from the 2J/ψ channel would need revision.","A direct lattice QCD calculation of X(6900) hadronic decay widths, using the same spin-2 currents, could independently decide between A-A and V-V without relying on the model's vertex shape."],"forward_implications":["If the A-A structure is correct, the two charm quarks in each diquark pair into an axial-vector state, which constrains the quantum numbers of the diquark and the overall tetraquark wavefunction.","The same coupling structure predicts X(6900)→2ηc with a width of 66–88 keV, offering an independent testable channel.","The vector–vector interpretation is disfavored, meaning any future model or lattice calculation should focus on the axial-vector diquark configuration.","The covariant quark model, with its compositeness-condition normalization, proves able to describe at least one fully heavy exotic hadron, encouraging its extension to other tetraquark candidates.","The computed widths are only weakly sensitive to the vertex size parameters, so the ordering between A-A and V-V is a robust feature within this framework."],"fun_headline_variants":["X(6900) likely axial-vector tetraquark from decay rates","Axial-vector beats vector in X(6900) tetraquark model","Decay to 2J/ψ exposes X(6900) as axial-vector state","Quark model: X(6900) axial-vector coupling fits data"],"cache_read_input_tokens":2304,"weakest_assumption_plain":"The whole comparison with experiment rests on the assumed Gaussian vertex shape (with scale around the X mass) and on the asserted cancellation of two of the three Feynman diagrams; if either is wrong, the A-A decay width could shift outside the measured range.","fun_headline_variants_meta":{"raw":{"variants":["X(6900) likely axial-vector tetraquark from decay rates","Axial-vector beats vector in X(6900) tetraquark model","Decay to 2J/ψ exposes X(6900) as axial-vector state","Quark model: X(6900) axial-vector coupling fits data"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.00058,"raw_usage":{"total_tokens":2547,"prompt_tokens":700,"completion_tokens":1847,"prompt_tokens_details":{"cached_tokens":256},"prompt_cache_hit_tokens":256,"prompt_cache_miss_tokens":444,"completion_tokens_details":{"reasoning_tokens":1759}},"tokens_in":444,"tokens_out":1847,"duration_ms":13630,"temperature":1.0,"reasoning_tokens":1759,"cache_read_input_tokens":256,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-03T14:57:52.751051+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Compute the crossed diagram contributions M_b + M_c explicitly (or replace the Gaussian vertex with a dipole form) and check whether the A-A width stays above the ~40 MeV lower edge of the data; if the cancellation fails or the shape change pulls the width below the measured values, the axial-vector–axial-vector conclusion would be weakened.","supporting_citations":[],"review_version":1}