{"id":"e45a1d9b-8bd0-497b-b078-f51352fd26df","arxiv_id":"2607.24446","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":3.5,"correctness_risk":"medium","formal_verification":"none","parameter_count":6,"one_line_summary":"In this U(1)_X model, κ_γ stays within ATLAS/CMS 1σ while charged Higgses are pushed above roughly 650–700 GeV and λ_13+λ_14 obey simple lower bounds set by the H–heavy-Higgs mixings.","lead":"A flavor-dependent U(1) extension of the Standard Model keeps the 125 GeV Higgs couplings to fermions and W/Z essentially Standard-Model-like, while new charged scalars can shift the diphoton rate. The paper maps which scalar couplings and charged-Higgs masses still match ATLAS and CMS κ_γ data at 1σ.","discovery_kind":"extension","skeptic_critique":{"model":"moonshotai/kimi-k3","headline":"The \"allowed\" quartic regions and the charged-Higgs mass floors are derived with only |λ|<4π perturbativity — vacuum-stability and unitarity cuts on the multi-scalar potential are never imposed, and the missing constraints act precisely where the floors come from.","rationale":"The reader already judged the paper CONDITIONAL with medium correctness risk, citing hand-fixed ε values, forced quartic equalities, and unreproducible scan detail. I agree with that verdict level and do not think the missing BFB/unitarity cuts raise it to REJECT: the loop calculation (Eqs. 43–45), the κ definitions, and the small-mixing approximations are conventional and internally consistent, and the neglected effects the reader flagged (O(ε²) corrections to κ_W,Z,f, O(εv/Λ) Yukawa corrections) are genuinely ≤10⁻³ for ε ≤ 10⁻², so they do not threaten the κ_γ consistency claim. The κ_γ=1 statement itself is safe. What is not safe without further work is the quantitative edge of the allowed quartic region and the derived mass floors, because the scan enforces only |λ|<4π while omitting the standard boundedness-from-below and unitarity constraints that are known to exclude large-negative portal-coupling corners in multi-scalar potentials — the very corner (λ13+λ14 ≈ 5−4π) that defines the paper's headline correlation. This is a checkable, not speculative, gap: the conditions are algebraic and can be applied post hoc to the existing scan. Hence UNCHANGED (remain CONDITIONAL), with the concrete test above as the condition that would either firm up or move the quoted floors. Agreement with the reader is partial: same overall posture, but the load-bearing weakness I identify is the absent theoretical constraint set rather than the scan simplifications per se.","tokens_in":12214,"tokens_out":3836,"duration_ms":150493,"concrete_test":"Take the surviving scan points underlying Figs. 1–2 and apply the standard copositivity BFB conditions for the two-Z2-odd-doublet potential (λ7,λ8>0; λ9+2√(λ7λ8)≥0; λ9+λ10+2√(λ7λ8)≥0; and the ϕ-portal/χ-portal analogues for λ11,λ13 vs √(λ1λ7), λ12,λ14 vs √(λ1λ8), λ15,λ17, λ16,λ18 vs the singlet self-couplings), plus tree-level unitarity on scalar 2→2 eigenvalues. Report (a) the fraction of Fig. 1 points that survive and (b) the recomputed charged-Higgs floors. If the floors move by more than ~50 GeV or the λ13+λ14 lower edge shifts, the headline numbers in the abstract are not robust to mandatory theoretical constraints.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The strongest claim is that κ_γ 1σ consistency yields (i) an allowed quartic region λ13+λ14 ≥ 5−4π (or ≥ −4π) and (ii) strengthened charged-Higgs floors m_H±1 ≥ 700 GeV etc. Both are scan outputs, and the scan (Sect. III.A) applies exactly one theoretical cut on the quartics: perturbativity |λ_ij| < 4π. But this is a four-doublet/singlet potential (Eqs. 10–11) with ~19 quartics, and the physically allowed region of such potentials is normally carved out by bounded-from-below/copositivity conditions coupling the portal quartics to the self-couplings (e.g., for the η–ρ subsystem λ9 + 2√(λ7λ8) ≥ 0, λ9+λ10 + 2√(λ7λ8) ≥ 0, and ϕ-portal conditions like λ11+λ13 + 2√(λ1λ7) ≥ 0), plus perturbative-unitarity limits on 2→2 scalar scattering that are typically much tighter than |λ|<4π for the eigenvalue combinations. None of these are mentioned anywhere in the paper. This is not a cosmetic omission: the λ13+λ14 lower edge (5−4π ≈ −7.6 for ε=10⁻³) sits at large negative combined portal coupling, exactly the corner where BFB conditions on the ϕ†ϕ(η†η) and ϕ†ϕ(ρ†ρ) directions bite hardest, and that edge is what generates the correlation plot and, through m²_H±1 ≈ M²_η + λ13 v²/2 and the cLFV window, feeds into the quoted mass floors. Similarly, the claim that \"the entire parameter space of λ17, λ18 is allowed\" (Fig. 1 right) is made without checking whether arbitrary λ17Λ1,2² contributions to M²_η, M²_ρ keep the potential stable and the R/I scalar masses (Eqs. 27–28) physical. If BFB/unitarity removes the negative-λ13−λ14 corner, the \"allowed region\" shrinks and the floors could shift in either direction; if it doesn't, the floors gain real support. Note this concern is complementary to, not identical with, the reader's flagged assumption (hand-fixed ε and forced quartic equalities): even with those simplifications accepted, the scan omits standard mandatory constraints.","agreement_with_reader":"partial"},"referee_report":{"model":"moonshotai/kimi-k3","summary":"The paper studies the 125 GeV Higgs boson in a flavor-dependent U(1)_X extension of the SM previously proposed by (overlapping) authors in Ref. [13]. The scalar sector contains the SM-like doublet phi, two Z_2-odd doublets eta, rho, and two singlets chi_{1,2}; the light CP-even state H mixes with two heavy states via small parameters epsilon_{1,2}. The authors show that, neglecting O(epsilon v/Lambda) Yukawa corrections, all tree-level couplings of H to SM fermions and gauge bosons equal their SM values (kappa_f = kappa_W = kappa_Z = kappa_g = 1), so the only modified coupling is kappa_gamma, which receives charged-scalar loop contributions from H1± and H2± (Eqs. 42–45). Imposing the 1-sigma ATLAS (1.01±0.06) and CMS (1.10±0.08) kappa_gamma values, together with flavor/collider windows imported from Ref. [13] (57.09 TeV ≤ 2√(Λ1²+9Λ2²) ≤ 78.92 TeV, m_H± ∈ [0.5,1.2] TeV, small charged mixing angle theta), and perturbativity |lambda_ij| < 4π, a numerical scan yields a correlation lambda_13+lambda_14 ≥ 5−4π (epsilon=10⁻³) or ≥ −4π (epsilon=10⁻²), no constraint on lambda_17,18, and strengthened charged-Higgs floors m_H±1 ≥ 700 GeV, m_H±2 ≥ 660 GeV (epsilon=10⁻³) and 650/680 GeV (epsilon=10⁻²).","tokens_in":12943,"tokens_out":7488,"duration_ms":266727,"significance":"The core calculation is standard and appears correct: the kappa_gamma width formula and loop functions (Eqs. 43–44) are the textbook ones, and kappa_f=kappa_V=1 follows cleanly from the charge assignment X_phi=0 and the singlet nature of chi_{1,2}. The paper delivers falsifiable, quantitative outputs — a sharp portal-coupling correlation and charged-Higgs mass floors in the 650–700 GeV range, testable in direct H± searches at the (HL-)LHC — which is a genuine strength. The work is, however, incremental in method: it applies a well-known one-loop constraint to a specific model whose entire external input (charge assignment, mass spectra, flavor/collider/DM windows) is taken from Ref. [13]. The robustness of the headline numbers is limited by the absence of vacuum-stability and unitarity cuts on the scalar potential, by an underspecified scan, and by the ad-hoc fixation of epsilon_{1,2}; these are addressable and do not challenge the qualitative conclusion that the model is currently consistent with kappa_gamma data.","major_comments":[{"comment":"The only theoretical cut applied to the quartics is |lambda_ij| < 4π. For this potential (Eqs. 10–11) the bounded-from-below conditions are substantially stronger: e.g., along the neutral phi–eta direction, V contains lambda_1|phi|^4 + lambda_7|eta|^4 + (lambda_11+lambda_13)|phi|²|eta|², requiring lambda_11+lambda_13 ≥ −2√(lambda_1 lambda_7); with the scan assumption lambda_11=lambda_13 this bounds lambda_13 ≥ −√(lambda_1 lambda_7), i.e. of order −0.4 to −1.6 depending on lambda_7 — far above the quoted edge lambda_13+lambda_14 ≥ 5−4π ≈ −7.6. The left edge of Fig. 1 (and hence the correlation plot and any floors that derive from it) is populated by points that generically destabilize the potential. Copositivity conditions for the eta–rho subsystem and perturbative-unitarity limits on 2→2 scalar scattering must be imposed and the figures redone, or the allowed regions must be explicitly l","section":"§III.A, Fig. 1 (left panel)"},{"comment":"The claim that 'the entire parameter space of lambda_17 and lambda_18 satisfies the 1-sigma bounds' is made without any physicality check. These couplings enter M²_eta = mu_4² + ... + lambda_17 Λ₂²/2 (and M²_rho similarly); with Λ₂ up to ~40 TeV and lambda_17 → −4π this is a shift of order −10⁷ GeV², which must be compensated by mu_4² to keep the charged-Higgs masses (Eqs. 30–31) and the R,I dark-scalar masses (Eqs. 27–28) positive and the Z_2-preserving vacuum a minimum. Since Fig. 2's mass floors are read off the same scan, the lambda_17,18 insensitivity conclusion and the floors are only as reliable as these unchecked tachyonic/vacuum constraints. A demonstration that all shown points correspond to physical spectra is needed.","section":"§III.A, Fig. 1 (right panel), Eqs. 27–31"},{"comment":"The strengthened floors m_H±1 ≥ 700 GeV etc. are a headline result, but the scan producing Fig. 2 is not described: no sampling ranges or priors for (Λ₁, Λ₂) within the 57–79 TeV window, no lambda ranges beyond |lambda|<4π, no point count, no statement of how the floors are extracted (scatter-cloud edge vs. analytic minimization). The floors are presumably edges of point clouds and thus sensitive to scan density; the inverted ordering between the two epsilon choices (700/660 at 10⁻³ vs. 650/680 at 10⁻²) is unexplained and hints at scan artifacts. The scan must be documented sufficiently for reproduction, and the floors shown to be stable under denser sampling.","section":"§III.A, Fig. 2"},{"comment":"The results are quoted for epsilon_{1,2} fixed by hand to 10⁻³ or 10⁻², yet Eqs. (18)–(19) define epsilon_{1,2} as functions of the chi-sector quartics and the very scales Λ_{1,2} being scanned. It is presumably possible to realize epsilon ~ 10⁻³–10⁻² by choice of lambda_2,...,lambda_6, lambda, but this is not demonstrated, and it is not shown that the natural size of epsilon over the scanned window is consistent with the chosen values (a parametric estimate epsilon ~ (coupling ratios)·v/Λ suggests 10⁻² is near the upper edge for Λ ~ 20–60 TeV). A consistency argument or a scan over a continuous epsilon range, rather than two discrete values, would make the per-epsilon mass floors well-defined.","section":"§II.B, Eqs. (18)–(19); §III.A"}],"minor_comments":[{"comment":"Eq. (46): m_Z = 91.67 GeV should read 91.19 GeV (PDG). Please correct.","section":"§III.A, Eq. (46)"},{"comment":"The abstract and conclusion state that the results satisfy 'constraints from ... dark matter studies,' but the scan imposes only the quark-flavor window, the cLFV mass window, and the small-theta condition; no dark-sector parameter (mu, lambda_19, M_eta,rho, relic density) enters the analysis. Either include the DM constraints explicitly or soften the claim.","section":"Abstract; §IV"},{"comment":"The kappa_gamma inputs are cited to the PDG review [19]; the primary ATLAS and CMS kappa-framework combination papers should be cited directly. Since the ATLAS (1.01±0.06) and CMS (1.10±0.08) central values differ by ~1 sigma, please state whether the intersection or the union of the two 1-sigma bands defines the allowed region in Figs. 1–2.","section":"§III, after Eq. (45)"},{"comment":"The notation t_{2ξ}, t_{2R,2I}, t_{2θ} (Eqs. 22, 29, 32) is used without definition; presumably t_{2x} ≡ tan(2x). Please define. Also 'mix through via a 2×2 matrix' (before Eq. 30) is garbled.","section":"§II.B"},{"comment":"The quark-flavor window is stated as 57.09 TeV ≤ 2√(Λ₁²+9Λ₂²) ≤ 78.92 TeV; given Eq. (36), m_Z' = 2 g_X z √(...), the factor of 2 suggests the bound is on m_Z'/(g_X z). Please state this explicitly and note the assumed g_X, z values inherited from Ref. [13].","section":"§III.A"},{"comment":"In Eq. (43) only the top-quark loop is retained among fermions; this is standard but should be stated. Relatedly, kappa_gamma as defined in Eq. (45) is a partial-width ratio; using it as a signal-strength proxy assumes the total width is SM-like. Given TeV-scale Z_2-odd masses, exotic decays such as H→R₁R₁ are kinematically closed — please state this assumption explicitly.","section":"§III, Eqs. (43)–(45)"},{"comment":"kappa_S is normalized to g_S^SM = −g²v/2, which is not an SM quantity (the SM has no charged scalar). Please clarify that this is merely a normalization convention and give the explicit mapping from g_HSS to the amplitude coefficient.","section":"§III, below Eq. (43)"},{"comment":"The numerical kappa_g formula (1.042 kappa_t² − ...) is quoted without a source; please cite the kappa-framework/Higgs-cross-section-working-group origin. In Figs. 1–2 the two epsilon choices are distinguished only by color (blue/red); a marker-style distinction would survive grayscale printing.","section":"§III; Figs. 1–2"}],"recommendation":"major_revision","confidential_remarks":"The manuscript is a competent but incremental application of a standard one-loop kappa_gamma constraint to a model taken wholesale from Ref. [13], by a largely overlapping author group; all external input windows (flavor, collider, DM) are imported from that single prior paper, so the standalone novelty is limited to the Higgs-coupling analysis itself. This is not grounds for rejection — the quantitative floors are useful targets for charged-Higgs searches — but the editor may wish to weigh the thin incremental layer over Ref. [13] against the journal's originality threshold. The required fixes (BFB/unitarity cuts, scan documentation) are standard and well within the authors' capability."},"author_rebuttal":null,"desk_editor":{"model":"grok-4.5","letter":"This is a clean follow-up to the authors’ own Ref. [13]. What is actually new is the one-loop κ_γ analysis, the H–H± couplings, and the resulting 1σ scatter regions for λ13–λ14 and the charged-Higgs masses. Everything else—the charge assignment, residual Z2, mass formulas, and the flavor/collider/DM windows—is imported.\n\nThey do the standard pieces correctly. κ_f = κ_W = κ_Z = 1 follows immediately from the charge assignments and the small-mixing approximation; the diphoton width and loop functions are textbook; and the numerical claim that κ_γ sits inside the ATLAS/CMS 1σ bands while obeying the prior cuts is credible on its face. The slightly stronger floors (m_H±1 ≳ 700 GeV, m_H±2 ≳ 660 GeV for ε = 10^{-3}) are a genuine, if modest, output.\n\nTwo soft spots matter in proportion. First, the scan freezes ε1,2 to two discrete values, forces λ11=λ13 etc. “for simplicity,” and drops the O(ε v/Λ) Yukawa corrections so that every tree-level κ is exactly 1. That is transparent but under-explored; the maps would shift if those equalities are relaxed. Second—and this is the stress-test point that holds up—the only theoretical cut on the quartics is |\\lambda| < 4π. The multi-scalar potential has the usual bounded-from-below and unitarity requirements on the portal combinations; none appear. The quoted lower edge λ13 + λ14 ≳ 5 − 4π sits precisely where those conditions bite, so the “allowed” corner and the mass floors that feed off it are not yet fully vetted. That is a real gap, not a fatal one; it is the sort of thing a referee will ask to be fixed or justified.\n\nCitation pattern is heavy on the group’s recent papers, which is expected for a model-specific follow-up and not itself a problem. No code or sampling details are supplied, so the figures are not regenerable.\n\nWho it is for: people already working on this U(1)_X construction or on charged-Higgs contributions to h → γγ. It will not move anyone outside that circle. It deserves a serious referee rather than a desk reject; the calculation is competent and the new numerical constraints are worth having once the vacuum-stability cuts are added. I would engage if I were already citing [13]; otherwise I would wait for the revised version.","headline":"Solid incremental κ_γ scan on the authors’ own U(1)_X model; new mass floors are real but rest on hand-fixed mixings and missing BFB/unitarity cuts.","tokens_in":13579,"tokens_out":642,"would_cite":false,"duration_ms":14034,"reading_group":"no","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":[],"pacs":[],"model":"grok-4.5","headline":"A flavor-dependent U(1) extension keeps the 125 GeV Higgs SM-like in couplings and matches LHC diphoton data while tightening charged-Higgs mass floors.","keywords":["flavor-dependent U(1)_X","SM-like Higgs","kappa_gamma","charged Higgs bosons","Higgs signal strengths","diphoton decay","beyond Standard Model"],"falsifier":"A future LHC or HL-LHC measurement of kappa_gamma (or of the charged-Higgs masses) that falls outside the regions allowed by the paper’s 1-sigma scan for the quoted epsilon values would rule out the claimed consistency.","tokens_in":12893,"feed_emoji":"⚛️","tokens_out":981,"duration_ms":19398,"temperature":0.7,"pith_summary":"The paper studies the light Higgs H in a flavor-dependent U(1)_X extension that already aims to explain three fermion generations, neutrino masses, dark matter, and flavor anomalies. It shows that H couples to SM fermions and to W and Z exactly as in the Standard Model, so the usual kappa modifiers for those channels stay equal to one. The only new handle is the one-loop H to gamma-gamma rate, which receives contributions from two new charged scalars. After scanning the relevant quartic couplings and mixing parameters under existing flavor, collider, and dark-matter bounds, the model still lands inside the ATLAS and CMS 1-sigma bands on kappa_gamma. The same requirement raises the lower limits on the charged-Higgs masses above the earlier literature values. A sympathetic reader cares because the construction remains viable under precision Higgs data without spoiling the mechanisms that motivated the extra U(1) in the first place.","feed_headline":"Flavor U(1) Higgs stays SM-like and matches LHC diphoton data","feed_subtitle":"New charged scalars keep kappa_gamma inside ATLAS/CMS 1-sigma while raising their mass floors above 650 GeV","key_machinery":"The kappa_gamma modifier built from the one-loop H to gamma-gamma width, with extra contributions proportional to the couplings g_H H1+ H1- and g_H H2+ H2- that arise from the SM-doublet–odd-doublet quartics and from the small mixings epsilon_1,2.","core_discovery":"In this flavor-dependent U(1)_X model the SM-like Higgs H has tree-level couplings to fermions and to W/Z identical to the Standard Model, while its one-loop diphoton rate (parameterized by kappa_gamma) remains consistent with ATLAS and CMS 1-sigma measurements once the new charged scalars and the small H–heavy-Higgs mixings are included; the same 1-sigma window strengthens the charged-Higgs mass floors to roughly 650–700 GeV.","pith_inferences":["Because kappa_f = kappa_W = kappa_Z = 1 by construction, any future deviation in those modifiers would immediately exclude the present charge assignment or the assumed VEV hierarchy.","The diphoton channel is currently the only sensitive Higgs probe of the new charged scalars; improved kappa_Zgamma data could supply an independent cross-check once experimental precision improves.","The same charged scalars that correct kappa_gamma also mediate lepton-flavor violation, so a joint fit of diphoton and LFV rates could shrink the surviving parameter space further."],"forward_implications":["Tree-level Higgs couplings to fermions, W and Z remain Standard-Model-like, so no tension is expected in those kappa channels.","Allowed quartic combinations must satisfy lambda_13 + lambda_14 greater than or equal to 5-4pi (or -4pi) according to the size of the mixings.","Charged-Higgs masses are pushed above approximately 650–700 GeV, stronger than earlier bounds from the same model.","The two heavy CP-even scalars and the new Z-prime remain at the multi-TeV scale set by the U(1)_X-breaking VEVs."],"fun_headline_variants":["Flavor U(1) SM-like Higgs keeps kappa_gamma in ATLAS/CMS 1-sigma","Charged scalars push mass floors to 650 GeV while Higgs stays SM-like","SM-like H in flavor U(1)_X matches LHC diphoton at one loop","New charged scalars keep H diphoton rate inside 1-sigma LHC bounds","Flavor-dependent U(1) Higgs couplings match SM and LHC kappa_gamma"],"cache_read_input_tokens":128,"weakest_assumption_plain":"The numerical maps rest on setting several pairs of quartic couplings equal for simplicity, fixing the heavy-Higgs mixings to two discrete tiny values, and dropping the small corrections to the fermion Yukawas so that all other kappa factors equal one exactly.","fun_headline_variants_meta":{"raw":{"variants":["Flavor U(1) SM-like Higgs keeps kappa_gamma in ATLAS/CMS 1-sigma","Charged scalars push mass floors to 650 GeV while Higgs stays SM-like","SM-like H in flavor U(1)_X matches LHC diphoton at one loop","New charged scalars keep H diphoton rate inside 1-sigma LHC bounds","Flavor-dependent U(1) Higgs couplings match SM and LHC kappa_gamma"]},"model":"grok-4.5","effort":"low","cost_usd":0.004721,"raw_usage":{"total_tokens":1342,"prompt_tokens":724,"num_sources_used":0,"completion_tokens":99,"cost_in_usd_ticks":47208000,"prompt_tokens_details":{"text_tokens":724,"audio_tokens":0,"image_tokens":0,"cached_tokens":256},"completion_tokens_details":{"audio_tokens":0,"reasoning_tokens":519,"accepted_prediction_tokens":0,"rejected_prediction_tokens":0}},"tokens_in":724,"tokens_out":99,"duration_ms":8534,"temperature":1.0,"reasoning_tokens":519,"cache_read_input_tokens":256,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-07-31T14:42:53.953380+00:00","model_set":{"reader":"grok-4.5"},"falsifier":"A future LHC or HL-LHC measurement of kappa_gamma (or of the charged-Higgs masses) that falls outside the regions allowed by the paper’s 1-sigma scan for the quoted epsilon values would rule out the claimed consistency.","supporting_citations":[],"review_version":1}