{"id":"0f90a81c-decc-4db9-b6a5-148c86d7fcd6","arxiv_id":"1908.09277","paper_version":2,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":6,"one_line_summary":"An extra U(1) gauge force with a small mass split between dark matter and a partner can give the right relic abundance while suppressing today's annihilation and direct detection signals.","lead":"This paper proposes dark matter models where a new U(1) force makes dark matter annihilate efficiently in the early universe but almost never interact today, explaining why direct detection experiments see nothing. The idea is applied to three specific U(1) extensions, with parameter regions that also touch the muon g-2 anomaly and the Hubble tension.","discovery_kind":"new_application","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The claimed suppression of present-day signals rests on a fine-tuned, radiatively unstable benchmark (λ4=0, sinα=1e-3) rather than on the gauge structure; a one-loop RGE check is needed.","rationale":"I read the paper as a model-building demonstration: in extra U(1) extensions, a complex scalar can freeze out via gauge coannihilation while its present annihilation and direct detection rates are suppressed through a small h−S coupling. The Boltzmann treatment and the cross-section formulas in the appendices are standard, and the three anomaly-free examples are worked out consistently. The reader's CONDITIONAL verdict is appropriate. My stress-test agrees with the reader that the weakest assumption is the fixed parameter choice that creates the suppression, but I would sharpen it: the central issue is not merely that the values are 'chosen' but that they are radiatively unstable in a way that directly feeds back into the observables the mechanism is designed to suppress. Setting λ4=0 and sinα=1e-3 without a protective symmetry means the benchmark is a tuned slice, and radiative corrections can generate an effective h−S coupling that exceeds the value needed to stay below the XENON1T and Fermi bounds. This does not make the paper wrong, but it makes the claimed viability conditional on a check the paper does not perform. I do not see an internal inconsistency in the derivation itself, so I would not reject the paper or mark it unverified; the conditionality is already captured by the reader's verdict. I therefore leave the verdict unchanged. My agreement is 'partial' because the reader also lists sub-GeV constraints, which I do not regard as equally load-bearing for the central claim; the radiative-stability issue is the decisive one.","tokens_in":17139,"tokens_out":28647,"duration_ms":322195,"concrete_test":"Take a representative benchmark from each model, e.g. the Z′ funnel point in Fig. 3 and the light Lμ−Lτ point mS=0.1 GeV. Implement the full scalar potential of Eq. (6) and compute the one-loop RGEs and, where relevant, the one-loop effective potential, evolving λ4, λ3, A, and the h−H mixing angle from the U(1) breaking scale to the electroweak scale. Then recompute σ_SI and (σv)_0 using the radiatively corrected h−S coupling from Eq. (17). If the induced coupling is large enough to raise σ_SI above the XENON1T limit or (σv)_0 above 10^-31 cm^3/s for the quoted masses, the benchmark is not radiatively stable and the claimed viable parameter regions shrink or vanish.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central mechanism is not the problem: the gauge coannihilation picture is coherent and the cross-section treatment is standard. The load-bearing issue is that the suppression of direct detection and of the present annihilation cross section is imposed by hand in the benchmark of Sec. III.B, (mP−mS, sinα, λ3, λ4)=(0.01mS, 1e−3, 1e−3, 0). Footnote 3 states this is chosen 'in order to have an efficient coannihilation and to have a tiny effective coupling between h and S,' but no symmetry enforces these values. In particular, λ4=0 is technically unnatural: the quartic λ4φ1†φ1|Φ|^2 is allowed by all gauge and global symmetries of the model, and even if set to zero at one scale it is generated by radiative corrections involving λ3, λ5, and the A term once φ2 acquires a VEV. The effective h−S coupling that controls both σ_SI and (σv)_0 is, from Eq. (17), λ4v cosα − (λ3v2 − √2A) sinα; the benchmark relies on this combination being tiny. A one-loop contribution to λ4 of order (λ3λ5)/(16π^2) can produce an h−S coupling comparable to or larger than the benchmark value, pushing σ_SI above the XENON1T/neutrino-floor bounds that motivated the mechanism and raising (σv)_0 above the quoted O(10^-31) cm^3/s. The paper provides no RGE or effective-potential analysis showing that these small parameters are radiatively stable. This is the softest point of the central claim; the rest of the construction is internally consistent.","agreement_with_reader":"partial"},"referee_report":{"model":"deepseek-v4-flash","summary":"The manuscript proposes a class of extra-U(1) extensions in which a complex scalar φ1 splits into two real scalars S and P after U(1) breaking. The lighter state S is the dark matter candidate; its thermal relic density is set by Z'-mediated coannihilation with P, while present-day annihilation and direct detection proceed through Higgs-boson exchange because Z'-mediated scattering is kinematically inelastic. The framework is applied to U(1)_{B-L}, U(1)_{(B-L)_3}, and U(1)_{L_μ-L_τ}, with parameter regions reproducing Ωh²≈0.1. The authors also discuss implications for direct detection, Fermi-LAT, the 47 Tucanae gamma-ray excess, the muon g−2 anomaly, IceCube, and the Hubble tension.","tokens_in":17506,"tokens_out":6341,"duration_ms":68299,"significance":"If the construction holds, it provides an existence proof for thermal-relic scalar dark matter whose present annihilation cross section can be as small as O(10^-31) cm^3/s while the freeze-out abundance is set by gauge coannihilation. This would resolve a common tension between thermal WIMPs and null direct-detection results. The manuscript's strengths are its explicit and internally consistent treatment: standard Boltzmann equations with coannihilation, transparent cross-section formulas in Appendices A and B, and concrete scans over three anomaly-free U(1) models with experimental constraints including LHC Z' bounds, BaBar data, neutrino trident bounds, and BBN limits. The main caveat is that the advertised suppression is not a consequence of the gauge structure alone; it is loaded into a benchmark whose radiative stability is not demonstrated.","major_comments":[{"comment":"The central suppression of direct and indirect signals is imposed by the benchmark choice (mP−mS, sinα, λ3, λ4) = (0.01 mS, 1e−3, 1e−3, 0). Footnote 3 states that these values are chosen 'in order to have an efficient coannihilation and to have a tiny effective coupling between h and S,' but no symmetry enforces them. In particular, λ4=0 is not protected: the quartic λ4 φ1†φ1 |Φ|² is allowed by all gauge and global symmetries of the scalar potential (6), and radiative corrections involving λ3, λ5, and the A term generate a nonzero λ4 once φ2 acquires a VEV. Since Eq. (17) shows the effective h−S coupling is λ4 v cosα − (λ3 v2 − √2 A) sinα, a one-loop shift δλ4 can alter σ_SI and (σv)_0 substantially, especially in the light-mass region. The manuscript contains no RGE or effective-potential analysis demonstrating that sinα=1e−3 and λ4=0 are radiatively stable at the scales relevant for freeze-out. Until this is quantified, the claim that the very suppressed present-day signals are a robust property of the models is not established.","section":"Sec. III.B, Eq. (17), footnote 3"},{"comment":"The light-DM region (mS ~ 0.1 GeV, mZ' ~ 0.01–0.1 GeV) is presented as an important consequence because it can simultaneously address the muon g−2 anomaly and relax the Hubble tension while evading Fermi-LAT bounds through annihilation into neutrinos. However, this region is not checked against sub-GeV direct detection via DM–electron scattering (generated by kinetic mixing), CMB bounds on s-wave annihilation into neutrinos or electromagnetic products, or constraints on the Z' decay through kinetic mixing. Footnote 7 explicitly defers the kinetic-mixing constraint to future work. Since the viability of this light region is used to motivate the phenomenology section, these missing constraints should be evaluated before the region is described as viable.","section":"Sec. VI.C, Figs. 7 and 8, footnote 7"}],"minor_comments":[{"comment":"The notation f_Tq and f_TG in Eq. (2) is not defined in the text; a brief definition or reference would improve readability.","section":"Sec. II.A, Eq. (2)"},{"comment":"Footnote 3 contains ungrammatical phrasing: 'We take as this in order to have...' should read 'We take these values in order to have...'.","section":"Sec. III.B, footnote 3"},{"comment":"The figure caption and text refer to several shaded regions ('light gray,' 'gray,' 'brown,' 'vermilion') that may be difficult to distinguish in grayscale print; explicit labels or line patterns would help.","section":"Sec. VI.B, Fig. 7"},{"comment":"The blue curve in Fig. 1 is called a 'theoretical interpretation' of the XENON1T bound; it should be stated more explicitly that this curve assumes the same Higgs-portal coupling controls both direct detection and present annihilation, an assumption that the model of Sec. III deliberately avoids.","section":"Sec. II.C, Fig. 1"},{"comment":"The Yukawa Lagrangian in Eq. (23) uses a notation where the Majorana mass term for N3_R involves φ2; it would be clearer to indicate the charge-conjugation contraction explicitly, as is done for the other N_R terms.","section":"Sec. V.A, Eq. (23)"}],"recommendation":"major_revision","confidential_remarks":"The paper is within scope for a hep-ph journal. The core mechanism (coannihilation through extra U(1) gauge interaction with inelastic Z' scattering) is coherent and the cross-section treatment is standard. The main concern is the radiative stability of the benchmark parameters that produce the very suppressed present-day signals; this is a well-defined technical issue that can be addressed with an RGE or effective-potential analysis, so I recommend major revision rather than rejection. If the stability analysis shows the benchmark is strongly tuned, the paper should be reframed as a proof-of-principle with explicit fine-tuning caveats."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"First thing to know: the paper's genuinely new move is the A(phi1 phi1 phi2^dagger) trilinear term in the scalar potential, which splits the complex scalar DM into S and P and turns the Z'-mediated scattering inelastic. That term is absent, by the authors' own account, in the earlier extra U(1) scalar DM papers they cite. As an existence proof for thermal relic DM with strongly suppressed present annihilation and direct detection, the mechanism works. The Boltzmann treatment, the coannihilation formulas in the appendices, and the direct-detection mapping are standard, and I found no internal error. The B-L, (B-L)_3, and L_mu-L_tau cases are worked through with the relevant collider and flavor constraints, and the comments on 47 Tucanae, muon g-2, and the Hubble tension are clearly framed as possibilities, not settled claims.\n\nThe soft spot is real and it is the one the reader flagged. The suppression that makes the scenario viable is loaded into the benchmark: Sec. III.B fixes (mP-mS, sin alpha, lambda3, lambda4) = (0.01 mS, 1e-3, 1e-3, 0), and footnote 3 says these are chosen precisely so that coannihilation is efficient and the h-S coupling is tiny. No symmetry enforces lambda4=0, and a one-loop contribution to lambda4 can easily produce an h-S coupling larger than the tiny value the plots rely on, pushing sigma_SI above XENON1T and raising (sigma v)_0. The paper gives no RGE or effective-potential check. I do not think this kills the paper; model-building papers routinely present benchmark points with small parameters. But the headline result is conditional on radiative stability, and a referee should ask for that check.\n\nMinor point: the sub-GeV L_mu-L_tau region is defended only against e+e- annihilation constraints; the authors note Z' then decays only to neutrinos, but they do not evaluate other sub-GeV probes, such as DM-electron scattering or CMB energy injection into neutrinos. Also, no data tables or code are provided for the figures, which makes independent numerical checking slower.\n\nWho is this for: anyone working on U(1) extensions, scalar dark matter, or the direct-detection/thermal-relic tension. It deserves a serious referee, with the radiative-stability question as the main request. My recommendation: send it out, ask for an RGE or symmetry argument, and do not treat the current benchmark as the end of the story.","headline":"A clean existence proof for inelastic extra U(1) scalar dark matter with suppressed present annihilation, worth refereeing, but the suppression depends on a technically unnatural benchmark that needs a radiative-stability check.","tokens_in":18119,"tokens_out":2327,"would_cite":true,"duration_ms":23572,"reading_group":"yes","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":[],"pacs":["95.35.+d"],"model":"deepseek-v4-flash","headline":"Thermal-relic scalar dark matter can have present-day annihilation below $10^{-31}\\,\\mathrm{cm}^3/\\mathrm{s}$: an extra $U(1)$ gauge force sets the relic density by coannihilation, while today's signals are suppressed Higgs-exchange…","keywords":["inelastic dark matter","scalar dark matter","coannihilation","extra U(1) gauge symmetry","thermal relic abundance","direct detection experiments","Z' boson","muon g-2 anomaly"],"falsifier":"Compute the one-loop quantum corrections to $\\lambda_4$ and $\\sin\\alpha$ in any of the three models: if the corrected effective Higgs-dark matter coupling pushes the spin-independent scattering cross section above the XENON1T limit (about $10^{-47}\\,\\mathrm{cm}^2$ for the relevant masses) anywhere along the relic-density contours, the claimed suppression is not realized. A cheaper direct check: for the sub-GeV $U(1)_{L_\\mu-L_\\tau}$ branch at $m_S\\simeq0.1\\,\\mathrm{GeV}$, compare the couplings required to reproduce $\\Omega h^2\\simeq0.1$ against dark matter-electron scattering and CMB bounds on annihilation into neutrinos, which the paper does not apply.","tokens_in":16863,"feed_emoji":"🌌","tokens_out":20400,"duration_ms":187404,"temperature":0.7,"pith_summary":"Direct detection null results force a scalar WIMP that annihilates through the SM Higgs to have a present-day annihilation cross section below about $10^{-31}\\,\\mathrm{cm}^3/\\mathrm{s}$ for most masses, far below the value needed to be a thermal relic. The paper shows that a complex scalar charged under an extra $U(1)$ gauge symmetry escapes this tension: the observed relic abundance is set by $Z'$-mediated coannihilation at freeze-out, while the same gauge boson cannot mediate elastic scattering off nuclei because its coupling is off-diagonal (inelastic), and present-day annihilation proceeds only through Higgs exchange with very small couplings. Viable parameter regions are identified in three anomaly-free models, $U(1)_{B-L}$, $U(1)_{(B-L)_3}$ and $U(1)_{L_\\mu-L_\\tau}$, and the flavored versions admit weak-scale or even sub-GeV dark matter with links to the muon $g-2$ anomaly, the Hubble tension and the 47 Tucanae gamma-ray excess. If the construction is correct, a thermal-relic WIMP with essentially no present-day annihilation signal is a concrete possibility rather than a fine-tuned exception.","feed_headline":"Dark matter can be a thermal relic, then go silent","feed_subtitle":"Coannihilation sets the relic abundance while present-day annihilation and scattering stay below detection.","key_machinery":"The load-bearing object is the nearly degenerate pair of real scalars $(S,P)$ that form a complex dark matter field charged under an extra $U(1)$, split in mass by the trilinear term $A(\\varphi_1\\varphi_1\\varphi_2^\\dagger+\\mathrm{h.c.})$ that couples it to the $U(1)$-breaking Higgs. That one term does two jobs: it sets the small splitting $m_P-m_S=0.01\\,m_S$, keeping the partner $P$ thermally abundant at freeze-out so that coannihilation (the pair-annihilation of $S$ together with $P$) through $s$-channel $Z'$ exchange sets the relic density, and it makes the gauge coupling off-diagonal, so that $Z'$-mediated dark matter-nucleus scattering is inelastic and kinematically suppressed. The remaining machinery is the effective thermally-averaged cross section that weights $S$ and $P$ annihilation channels by their equilibrium abundances (used to integrate the Boltzmann equation), and the $\\lambda_4$, $\\sin\\alpha$-suppressed Higgs-exchange couplings that set both the present annihilation rate and the elastic scattering cross section.","core_discovery":"The central claim is that freeze-out and the present-day signals of scalar dark matter can be carried by different interactions, decoupling the relic abundance from direct and indirect detection. In the construction, the dark sector is a complex scalar $\\varphi_1$ with charge $+1$ under a new $U(1)$, plus a Higgs field $\\varphi_2$ with charge $+2$ that breaks the symmetry; after symmetry breaking, dark matter is the lighter real component $S$, with a heavier coannihilating partner $P$ whose mass splitting is generated by the trilinear term $A(\\varphi_1\\varphi_1\\varphi_2^\\dagger+\\mathrm{h.c.})$. The gauge boson couples only to the off-diagonal current $Z'^\\mu((\\partial_\\mu S)P - S\\partial_\\mu P)$, so $Z'$-mediated scattering off nucleons is inelastic and inert once the splitting exceeds the recoil energy, and the relic density is fixed by coannihilation $SP\\to f\\bar f$ for $m_{Z'}>m_S$ or $SS\\to Z'Z'$ for $m_{Z'}