Pith. sign in

REVIEW 3 major objections 4 minor 71 references

Naturally resonant two-mediator model of self-interacting dark matter with decoupled relic abundance

T0 review · 3 major / 4 minor · reviewed 2026-08-06 · deepseek-v4-flash

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗

Pith's one-line read Two dark-sector mediators, one light for halo self-interactions and one heavy resonance at twice the dark-matter mass for freeze-out, satisfy both the relic abundance and galactic constraints, with a predicted 1.2 TeV signal.

desk verdict The benchmark that carries the paper violates the model's own mixing relations (θ≈58° vs |θ|≪1), and the direct-detection prediction is off by a factor 1000 due to a unit conversion error; the EFT construction is solid but the advertised island of viability sits outside the defined model. read the letter →

arxiv 2506.22997 v3 pith:SZ6KN3TB submitted 2025-06-28 hep-ph astro-ph.COgr-qc

Martin Drobczyk This is my paper
classification hep-phastro-ph.COgr-qc
keywords self-interactingdarkmatterresonantannihilationthermalfreeze-outSommerfeldenhancementtwo-mediatorsectorcompositedirectdetectiondensity-responsiveenergy
verification ladder T0 review T1 audit T2 compute T3 formal

The pith

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The reading

Minimal self-interacting dark matter models face a quantitative conflict: the coupling that produces the observed relic abundance through thermal freeze-out does not produce the self-scattering strength that dwarf galaxies require, and vice versa. The paper argues that a second mediator dissolves this conflict by giving the two jobs to two different particles. A light scalar $\phi$ with a mass near 15 MeV provides the velocity-dependent attractive Yukawa self-interactions that form cored halos, while a heavy scalar $\Phi_h$ tuned to the threshold $m_{\Phi_h}\approx 2m_\chi$ boosts the annihilation rate during freeze-out by a factor of about 143 through an $s$-channel resonance. A numerical scan finds a narrow island of viability, with a representative benchmark at $m_\chi=600$ GeV, $m_\phi=15$ MeV, and $m_{\Phi_h}=1201$ GeV that matches the measured relic density and gives $\sigma_T/m_\chi$ around 1 to 0.1 cm$^2$/g at dwarf-galaxy velocities while falling four orders of magnitude at cluster scales. If the model is right, it predicts a narrow top-pair resonance near 1.2 TeV within reach of the High-Luminosity LHC and a spin-independent direct-detection cross-section near $7\times10^{-48}$ cm$^2$.

What carries the argument

The machinery is a scalar pair emerging from one complex field whose potential is spontaneously broken: a heavy radial mode $\Phi_h$ with mass $\sqrt{2\lambda}\,v_s$ at the TeV scale, and a light pseudo-Nambu-Goldstone boson $\phi$ at the MeV scale whose small mass is protected by an approximate shift symmetry, mixed by a tiny CP-violating angle $\theta$. Because the dark fermion $\chi$ receives its mass from the same vacuum expectation value, the resonance condition $m_{\Phi_h}\approx2m_\chi$ reduces to the coupling relation $\lambda\approx y_f^2$, and the small detuning $\delta=m_{\Phi_h}/(2m_\chi)-1\simeq8.3\times10^{-4}$ is shown to be radiatively stable. Annihilation proceeds through the $s$-channel Breit-Wigner pole with a total enhancement factor $S_{\rm total}\simeq143$ that includes Sommerfeld enhancement from the light mediator; self-interactions are computed non-perturbatively by solving the Schrödinger equation for the attractive Yukawa potential in partial waves.

What would settle it

A dedicated search for a narrow resonance in the $t\bar t$ invariant-mass spectrum near 1.2 TeV with the full High-Luminosity LHC dataset: if no peak appears at the predicted production rate of order 1–10 fb, the benchmark's resonant freeze-out mechanism is ruled out. As an independent check, a next-generation direct-detection experiment reaching $\sigma_{\rm SI}\sim10^{-48}$ cm$^2$ that sees no events near $m_\chi\simeq600$ GeV would falsify the portal structure the benchmark requires.

Watch

Extended reading notes

Core claim

The central claim is that the long-standing tension between the thermal relic abundance and the self-interaction requirement is not intrinsic to self-interacting dark matter but a symptom of giving one mediator both jobs. The paper first quantifies the failure of the minimal one-mediator model—at $m_\chi=100$ GeV and $m_\phi=20$ MeV, the coupling that gives $\sigma_T/m_\chi=1$ cm$^2$/g at 30 km/s overproduces dark matter by a factor of 2.5—and then shows that adding a heavy scalar $\Phi_h$ near the $2m_\chi$ threshold restores consistency without changing late-time self-interactions. The proof of concept is a non-empty intersection of the relic-density band and the self-interaction band in the $(m_\phi, y_\chi)$ plane, a narrow island spanning roughly $m_\phi\in[12,18]$ MeV and $y_\chi\in[0.28,0.32]$ at $m_\chi=600$ GeV once the heavy-sector parameters are fixed by the relic abundance alone.

Load-bearing premise

The model only works if the light mediator is completely silent to quarks and couples to electrons through an extremely small constant, $c_e\simeq5\times10^{-11}$, chosen by hand; that input—not the resonance—is what keeps the model within direct-detection and Big Bang nucleosynthesis bounds.

Editorial extensions

If this is right

  • A narrow scalar resonance at about 1.2 TeV with total width $\Gamma_{\Phi_h}\simeq0.17$ GeV and branching ratio to $t\bar t$ of 99.85% should appear as a peak in the top-pair invariant mass spectrum, with a production rate of order 1–10 fb at 14 TeV that puts it within the reach of the 3000 fb$^{-1}$ High-Luminosity LHC dataset.
  • The spin-independent direct-detection cross-section is pinned near $\sigma_{\rm SI}\simeq6.7\times10^{-48}$ cm$^2$ for the 600 GeV benchmark, just below the current leading limit and inside the projected reach of next-generation liquid-xenon experiments.
  • The self-interaction cross-section has a sharp, specific velocity dependence—0.96 cm$^2$/g at 10 km/s, 0.11 cm$^2$/g at 30 km/s, and $9.5\times10^{-5}$ cm$^2$/g at 1000 km/s—so halo and cluster observations can test the model independently of collider results.
  • Because the resonance operates only at freeze-out velocities, today's annihilation rate is suppressed by a factor of about 143 relative to the canonical thermal value, placing indirect gamma-ray signals safely below current dwarf-galaxy limits.
  • The viable region scales predictably across $m_\chi\in[200,1000]$ GeV as $m_\phi\propto m_\chi^{0.83}$ and $y_\chi\propto m_\chi^{0.51}$, so a detection at any mass fixes the entire family of allowed benchmarks.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

  • The paper leaves the cosmology of its density-responsive dark-energy sector uncomputed; a dedicated Boltzmann analysis of the $\rho_\Phi(X)$ evolution would show whether the mild $w(a)$ deviation it permits also shifts the predicted value of $S_8$, a connection the paper raises but does not quantify.
  • The quark-silent, leptophilic structure of $\phi$ occupies a narrow experimental corridor: with $m_\phi=15$ MeV and $c_e\simeq5\times10^{-11}$ the mediator lives about 0.44 s, so tightened nucleosynthesis bounds or new light-scalar beam-dump searches would test this imposed input rather than the resonance mechanism itself.
  • If the composite SU(3)$_H$ completion is taken seriously, the confinement transition at $\Lambda_H\simeq2.5$ TeV should radiate a stochastic gravitational-wave background; estimating its amplitude would convert the optional UV story into a directly testable signature.
  • The same decoupling logic—a threshold resonance setting the abundance while a light state controls halo physics—suggests that the one-mediator tension is a general structural feature of self-interacting dark matter, not a peculiarity of the Yukawa setup studied here.
Share X Bluesky LinkedIn Reddit HN

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Referee Report

3 major / 4 minor

Summary. The paper proposes a two-mediator dark-sector EFT in which a light scalar φ (MeV scale) mediates velocity-dependent self-interactions while a heavy scalar Φ_h (TeV scale) near 2m_χ resonantly enhances annihilation during thermal freeze-out, decoupling early-universe relic production from late-time halo dynamics. A numerical scan is claimed to identify a narrow viable island, with a benchmark m_χ = 600 GeV, m_φ = 15 MeV, m_Φh = 1201 GeV, y_χ = 0.30, g_Y1^DM = 0.190 that reproduces Ωh² ≈ 0.120 and σ_T/m_χ ~ 0.1–1 cm²/g at dwarf velocities. The paper also presents LHC, direct-detection, BBN, and indirect-detection signatures, and an optional composite SU(3)_H completion.

Significance. The conceptual mechanism—separating the mediator responsible for SIDM from the mediator responsible for freeze-out—is well motivated and could address a genuine tension in minimal SIDM models. The paper is unusually explicit about its model assumptions and provides extensive appendices, including a radiative-stability check in Appendix G and a detailed scan description in Appendix D. However, the central benchmark is not a point of the model as defined by the paper's own coupling relations in Section 2.2, and the direct-detection prediction contains a factor-1000 unit conversion error. Because the advertised proof-of-concept and one of the headline observables are invalidated, the significance of the paper as it stands is conditional at best.

major comments (3)
  1. [§2.2, Eqs. (7)–(9); Appendix A.3; Table 1] The Table 1 benchmark is inconsistent with the model's own coupling relations. Equations (9) and (A.9) give y_χ = (y_f/√2) sinθ and g_Y1^DM = (y_f/√2) cosθ, so tanθ = y_χ/g_Y1^DM = 0.30/0.190 = 1.58, i.e. θ ≈ 58°, in direct contradiction to the assumed |θ| ≪ 1 in Eq. (5) and (A.8). Such a mixing angle is also impossible in the mass matrix (A.34)–(A.37) with m_φ = 15 MeV and m_Φh = 1201 GeV: determinant positivity requires |θ| ≲ m_φ/m_Φh ≈ 1.2×10⁻⁵. The numerical scan in Appendix D.2 treats y_χ and g_Y1^DM as independent parameters and never enforces Eq. (9), so the advertised island of viability may lie entirely outside the model defined in Section 2.2.
  2. [Appendix C.4, Eq. (C.8)] The unit conversion in Eq. (C.8) is wrong by a factor of 1000: 1 GeV⁻² = 0.3894×10⁻²⁷ cm² (0.3894 mb), not 0.3894×10⁻²⁴ cm². Therefore the predicted spin-independent cross-section is σ_SI ≈ 6.7×10⁻⁵¹ cm², not 6.7×10⁻⁴⁸ cm². This removes the claimed direct-detection signal from the reach of next-generation experiments and invalidates the corresponding statements in the abstract, Section 5.4, and Section 7.
  3. [§5.5] The collider production rate is quoted without any calculation. Section 5.5 states an 'O(1–10) fb' rate for σ(pp→Φ_h→ttbar) at √s = 14 TeV, but no gluon-fusion cross-section formula, PDF choice, acceptance estimate, or background analysis is given; Eq. (33) is merely schematic. Since 'within HL-LHC reach' is one of the paper's headline falsifiable predictions, this rate must be substantiated with an actual calculation or a quantitative reference.
minor comments (4)
  1. [Throughout] There are numerous typographical and grammatical errors, including 'identifed' (Section 1), 'Altough' (Section 4.2), 'is are be specified' (Section 2.2), and inconsistent rendering of 'Schrödinger' in Appendix F; a careful proofread is needed.
  2. [§5.3, Appendix B] The claim that φ 'never thermalizes' is based on Γ/H ~ 0.05–0.15 at T ≈ m_φ, which is not far below unity; the authors should quantify the actual freeze-in abundance and justify the non-thermalization statement with the full Boltzmann yield rather than a rough ratio.
  3. [§5.5, Appendix C.4] The terminology 'quark-only portal' for Φ_h and 'quark-silent' for φ should be clearly separated at first use to avoid an apparent contradiction: the heavy scalar couples to quarks while the light scalar is leptophilic and quark-silent at tree level.
  4. [§4.1, Appendix D.2] The two-stage scan explicitly tunes the heavy-sector parameters to reproduce Ωh² and selects y_χ and m_φ along the SIDM band; this is a legitimate parameter search, but the language of 'predictive island' should be softened to reflect that the observables are used as inputs to locate the benchmark.

Circularity Check

3 steps flagged · score 6.0 of 10

The benchmark's relic-density consistency is fitted rather than predicted: Section 4.1 tunes the heavy sector to the relic target, and the UV resonance condition is manufactured by choosing k_Phi=0.48; the benchmark also violates the model's own Eq. (9) mixing relation, though direct-detection and LHC signatures retain independent content.

  1. fitted input called prediction [Section 4.1 (Numerical methodology, step ii) and Section 4.2 (Viable parameter space)]
    "Along this SIDM solution band, we then adjusted the heavy sector parameters, primarily the detuning δ = (mΦh −2mχ)/(2mχ) and the coupling gDMY1, to reproduce the observed relic abundance, Ωh2 = 0.120±0.001."

    The paper presents the green intersection in Figure 1 as a 'non-trivial outcome' and 'central proof of concept', but the entire blue relic-density band is generated by tuning δ and gDMY1 to the Planck target. Because the heavy sector is adjusted independently for each light-sector point, the benchmark's Ωh²=0.119 is a restatement of the fit target, not a derived prediction. The 'prediction' of the correct relic density reduces by construction to the fitted input.

  2. other [Section 2.2, Eqs. (5) and (9); Appendix A.3, Eqs. (A.34)-(A.37); Table 1]
    "yχ ≃ yf√2 sinθ, gDMY1 ≃ yf√2 cosθ."

    Table 1 sets yχ=0.30 and gDMY1=0.190. Eq. (9) then forces tanθ=yχ/gDMY1=1.58, i.e. θ≈58°, in direct contradiction to the |θ|≪1 rotation (Eq. (5), Eq. (A.35)) used to define the light mediator and the small-θ mass matrix. Inserting this angle into Eqs. (A.34)-(A.37) requires |m_sa^2|≈0.5 m_s^2, giving a negative determinant m_a^2 m_s^2 − (m_sa^2)^2 and hence no positive-definite spectrum for m_phi=15 MeV, m_Phi=1201 GeV. The advertised consistency of the benchmark is therefore not an output of the paper's own Yukawa/mixing relations; yχ and gDMY1 are free scanned inputs (Appendix D.1), so the central point is fitted outside the model rather than derived from it.

1 more flagged steps
  1. fitted input called prediction [Appendix H.3, Eq. (H.5); Section 6.2]
    "Lattice studies of near-conformal theories suggest kΦ∈[0.5,0.7] [38]. We adopted a value of kΦ = 0.48, which lies at the conservative lower edge of this range and yields the desired mΦh ≈1.2 TeV."

    The claimed dynamical origin of the resonance condition is obtained by choosing the coefficient kΦ=0.48 so that mΦh≈1.2 TeV after mχ≈600 GeV and ΛH≈2.5 TeV have been set. Eq. (H.5) then outputs mΦh/mχ≈2.0 by dividing this chosen coefficient by Nc/4π; the 'prediction' is the input. The quoted lattice range [0.5,0.7] does not justify the adopted value 0.48, which lies below it.

full rationale

The core DM benchmark is partially circular in the sense of the rubric: the relic-density constraint is enforced by tuning δ and gDMY1 (Section 4.1), and the self-interaction target is enforced by selecting mϕ and yχ on the SIDM band (Appendix D.2). The 'island of viability' is therefore a constructed intersection of two fitted constraints rather than a prediction of the model. In addition, the benchmark violates the model's own mixing relations: Eq. (9) with yχ=0.30 and gDMY1=0.190 gives θ≈58°, contradicting the |θ|≪1 assumption that underlies the PNGB light mediator and making the scalar mass matrix (A.34)-(A.37) non-positive-definite for mϕ=15 MeV, mΦh=1201 GeV. The numerical scan in Appendix D.1 indeed treats the two DM couplings as independent, so the advertised benchmark lives outside the EFT defined in Section 2.2. The optional SU(3)_H UV completion is not self-citation-load-bearing for the main DM claims: its resonance condition is manufactured by choosing kΦ=0.48, but the direct-detection cross-section, LHC resonance rate, and cluster-scale velocity suppression are genuine outputs from the chosen inputs and retain independent content. Ref. [9] is explicitly optional and does not carry the central argument. Overall score 6: several 'predictions' reduce by construction, while other signatures remain independent.

Assumptions & free parameters 8 free parameters · 5 assumptions · 4 invented entities

The central phenomenological claim rests on parameters that are fitted to the two target observables (relic density and self-interaction cross-section). The SU(3)_H UV completion adds more free choices. No machine-checked proof or archival code is provided.

free parameters (8)
  • Heavy resonance mass m_Phi_h = 1201 GeV
    Tuned so that the detuning delta = 8.3e-4 puts the resonance in the thermally effective region to reproduce the observed relic density (Section 4.1).
  • DM-heavy scalar coupling g_Y1 = 0.190
    Adjusted together with m_Phi_h to match Omega h^2 = 0.120 in the micrOMEGAs scan.
  • DM-light mediator coupling y_chi = 0.30
    Chosen within the viable band (0.28-0.32) to give the desired self-interaction cross-section of about 1 cm^2/g at dwarf velocities.
  • Light mediator mass m_phi = 15 MeV
    Selected within the viable band (12-18 MeV) to match the SIDM target while keeping the Yukawa potential range appropriate.
  • SM coupling of heavy scalar g_h,SM = 0.052
    Chosen to keep the direct detection cross-section below the LZ limit and to set the LHC production rate; not derived from the EFT.
  • Electron coupling of phi c_e = 5e-11
    Adopted so that phi decays before BBN (tau ~ 0.44 s); the value is picked in the range allowed by BBN constraints.
  • Composite scalar meson coefficient k_Phi = 0.48
    In the SU(3)_H UV completion, chosen at the lower edge of the lattice-informed range to yield m_Phi_h/m_chi ~ 2 (Appendix H.3).
  • Cosmological anomalous dimension gamma_cosmo = 0.50 +/- 0.05
    Inferred from all-orders estimates and lattice trends; chosen to connect the Planck scale to the meV dark energy scale, not derived in this paper.
assumptions (5)
  • standard math Standard thermal freeze-out formula for the relic density (Eq. 15) applies.
    Used in Section 3.1 and the micrOMEGAs calculation; standard cosmology is assumed.
  • domain assumption Sommerfeld enhancement factorizes with the Breit-Wigner resonant cross-section.
    Invoked in Section 4.1 and Appendix E.2, citing Beneke et al. [8]; this factorization is central to the combined enhancement calculation.
  • domain assumption The non-perturbative Yukawa scattering routines in micrOMEGAs correctly compute the momentum-transfer cross-section.
    The SIDM cross-sections in Section 5.2 and Appendix F rely entirely on this numerical tool.
  • ad hoc to paper The light mediator phi is quark-silent at tree level.
    Introduced in Appendix C.4 to avoid the m_phi^-4 enhancement of direct detection; this is an input, not a consequence of the symmetries.
  • standard math A global U(1)_chi symmetry ensures the stability of the dark matter fermion chi.
    Postulated in Section 2.1; standard practice in simplified DM EFTs.
invented entities (4)
  • SU(3)_H hidden gauge theory with N_f = 10
    purpose: Optional UV completion claimed to generate the composite DM spectrum, the resonance condition, and the anomalous dimension for dark energy.
    No experimental handle; relies on lattice and all-orders estimates for near-conformal dynamics.
  • Dark baryon chi (composite DM)
    purpose: Identified as the dark matter fermion in the composite picture.
    Hypothetical bound state with no independent evidence outside the model.
  • Dark scalar meson Phi_h (composite resonance)
    purpose: Identified as the heavy scalar resonance mediating resonant annihilation.
    Mass ratio to baryon tuned via k_Phi to match the EFT requirement.
  • Auxiliary non-propagating scalar Phi for dark energy
    purpose: Generates the density-responsive dark energy contribution rho_Phi(X).
    An algebraic degree of freedom in the EFT; no propagating mode and no direct observable.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Naturally resonant two-mediator model of self-interacting dark matter with decoupled relic abundance." pith.science (2026). https://pith.science/paper/SZ6KN3TB

@misc{pith2026250622997,
  author       = {Pith},
  title        = {Pith review of: Naturally resonant two-mediator model of self-interacting dark matter with decoupled relic abundance},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/SZ6KN3TB}},
  note         = {Machine review of arXiv:2506.22997}
}
abstract

We propose a minimal, fully thermal mechanism that resolves the long-standing tension between achieving the observed dark-matter relic abundance and explaining the astrophysical signatures of self-interactions. The framework introduces two mediators: a light scalar $\phi$ (MeV scale) that yields the required, velocity-dependent self-interactions, and a heavy scalar resonance $\Phi_h$ (TeV scale) with mass $m_{\Phi_h}\!\approx\!2m_\chi$ that opens an $s$-channel resonant annihilation during freeze-out. This clearly decouples early-universe annihilation from late-time halo dynamics. A detailed numerical analysis identified a narrow predictive island of viability. A representative benchmark with $m_\chi\!=\!600$~GeV, $m_\phi\!=\!15$~MeV, and $m_{\Phi_h}\!\simeq\!1.2$~TeV reproduces the relic density and yields $\sigma_T/m_\chi\sim 0.1$--$1~\mathrm{cm}^2\!/\mathrm{g}$ at dwarf-galaxy velocities while satisfying cluster bounds. The model makes sharp, testable predictions: a narrow $t\bar t$ resonance near $1.2$~TeV within HL-LHC reach, and a spin-independent direct-detection signal $\sigma_{\rm SI}\!\sim\!7\times10^{-48}\,\mathrm{cm}^2$ within next-generation sensitivity. As an optional UV completion, we show that walking $\mathrm{SU}(3)_H$ gauge theory with $N_f=10$ naturally realizes the near-threshold relation $m_{\Phi_h}\!\approx\!2m_\chi$ and can furnish an effective anomalous dimension $\gamma\!\approx\!0.5$ which underlies a density-responsive dark-energy sector, suggesting a unified origin for the dark sector.

Figures

Figures reproduced from arXiv: 2506.22997 by the authors.

Figure 1
Figure 1. Viable parameter space for self-interacting dark matter with resonant annihilation in the (mϕ, yχ) plane for a fixed dark matter mass of mχ = 600 GeV. The blue band shows the region satisfying the relic density constraint. The red contours define the target region for self-interactions (0.1 < σT /mχ < 10 cm2/g at v = 30 km/s). The green shaded region marks the intersection where both constraints are simultaneously s… view at source ↗
Figure 2
Figure 2. Scaling relations in the viable parameter space. Left: Light mediator mass versus dark matter mass. Right: Yukawa coupling versus dark matter mass. Green bands show the full viable region satisfying all constraints, circles indicate central values at discrete masses from our scan, and the red star marks our benchmark point. The dashed lines show the best-fit power laws mϕ ∝ m0.83 χ and yχ ∝ m0.51 χ . that this mass … view at source ↗
Figure 3
Figure 3. Resonant enhancement of dark matter annihilation. Top panel: Relic density Ωh 2 as a function of the resonance parameter δ = (mΦh /2mχ − 1). The solid blue line shows the full micrOMEGAs calculation including thermal and Sommerfeld effects. The gray band indicates the observed Planck 2018 value [1]. The red circle marks our benchmark point at δ ≈ 8.3 × 10−4 , yielding the correct relic abundance. Bottom panel: The c… view at source ↗
Figures from the paper (1 more)
Figure 4
Figure 4. Figure 4: Velocity-dependent self-interaction cross-section for our benchmark SIDM model. The solid green line shows σT /mχ for the best-fit parameters (mϕ = 15 MeV, yχ = 0.30), calculated using micrOMEGAs. Dashed and dotted lines illustrate the effect of varying mϕ and yχ. Colo…

Discussion (0). Sign in to comment.

Reference graph

Works this paper leans on

71 extracted references · 59 canonical work pages

  1. [1]

    Astrophys.641A6 [Erratum: Astron.Astrophys

    Aghanim Net al.(Planck) 2020Astron. Astrophys.641A6 [Erratum: Astron.Astrophys. 652, C4 (2021)]

    work page 2021

  2. [2]

    Bullock J S and Boylan-Kolchin M 2017Ann. Rev. Astron. Astrophys.55343–387

    work page

  3. [3]

    Rept.7301–57

    Tulin S and Yu H B 2018Phys. Rept.7301–57

    work page

  4. [4]

    Spergel D N and Steinhardt P J 2000Phys. Rev. Lett.843760–3763

    work page

  5. [5]

    Kaplinghat M, Tulin S and Yu H B 2016Phys. Rev. Lett.116041302

    work page

  6. [6]

    Tulin S, Yu H B and Zurek K M 2013Phys. Rev. D87115007

    work page

  7. [7]

    Oman K Aet al.2015Mon. Not. Roy. Astron. Soc.4523650–3665

    work page

  8. [8]

    Beneke M, Lederer S and Urban K 2023Phys. Lett. B839137773

    work page

Show all 71 references
  1. [9]

    Drobczyk M 2025arXiv:2506.12464(Preprinthttps://arxiv.org/abs/2506.12464)

  2. [10]

    Weinberg S 1996The Quantum Theory of Fields. Vol. 2: Modern Applications(Cambridge: Cambridge University Press)

  3. [11]

    Burgess C 2007Annual Review of Nuclear and Particle Science57329–362 ISSN 1545-4134

  4. [12]

    Khoury J and Weltman A 2004Phys. Rev. Lett.92071104 (Preprintastro-ph/0309300)

  5. [13]

    Hinterbichler K and Khoury J 2010Phys. Rev. Lett.104231301 (Preprint1001.4525)

    arXiv

  6. [14]

    Patt B and Wilczek F 2006 ArXiv:hep-ph/0605188 (Preprinthep-ph/0605188)

    2006 arXiv

  7. [15]

    Schabinger R M and Wells J D 2005Phys. Rev. D72093007 (Preprinthep-ph/0509209)

  8. [16]

    Goldstone J 1961Nuovo Cim.19154–164

  9. [17]

    Rev.127965–970 Naturally resonant two-mediator SIDM model with decoupled relic abundance41

    Goldstone Jand Salam A and Weinberg S 1962Phys. Rev.127965–970 Naturally resonant two-mediator SIDM model with decoupled relic abundance41

  10. [18]

    Gell-Mann M and Levy M 1960Nuovo Cim.16705

  11. [19]

    ’t Hooft G 1980NATO Sci. Ser. B59135–157

  12. [20]

    Gubser S S and Peebles P J E 2004Phys. Rev. D70123510 (Preprinthep-th/0402225)

  13. [21]

    Rockafellar R 1970Convex Analysis(Princeton University Press) ISBN 9781400873173

  14. [22]

    Kolb E W and Turner M S 1990The Early Universe(Frontiers in Physicsvol 69) (Redwood City, CA: Addison-Wesley)

  15. [23]

    Feng J L, Kaplinghat M and Yu H B 2010Phys. Rev. Lett.104151301

  16. [24]

    Arkani-Hamed N, Finkbeiner D P, Slatyer T R and Weiner N 2009Phys. Rev. D79015014

  17. [25]

    J.462563–575

    Navarro J F, Frenk C S and White S D M 1996Astrophys. J.462563–575

  18. [26]

    J.149180

    Oh S Het al.2015Astron. J.149180

  19. [27]

    Flores R A and Primack J R 1994Astrophys. J. Lett.427L1–L4

  20. [28]

    Zavala J, Vogelsberger M and Walker M G 2013Mon. Not. Roy. Astron. Soc.431L20–L24

  21. [29]

    Elbert O D, Bullock J S, Garrison-Kimmel S, Rocha M, Onorbe J and P A H G 2015Mon. Not. Roy. Astron. Soc.45329–37

  22. [30]

    J.606819–824

    Markevitch Met al.2004Astrophys. J.606819–824

  23. [31]

    Harvey D, Massey R, Kitching T, Taylor A and Tittley E 2015Science3471462–1465 ISSN 1095-9203

  24. [32]

    Alguero Get al.2024Comput. Phys. Commun.299109133

  25. [33]

    Backovic M, Kr¨ amer M, Maltoni F, Martini A, Mawatari K and Pellen M 2015 FeynRules model database: DMsimphttps://cp3.irmp.ucl.ac.be/projects/feynrules/wiki/DMsimp accessed: 5 June 2025

    2015

  26. [34]

    Rocha M, Peter A H G, Bullock J S, Kaplinghat M, Garrison-Kimmel S, Onorbe J and Moustakas L A 2013Mon. Not. Roy. Astron. Soc.43081–104

  27. [35]

    Manohar A and Georgi H 1984Nucl. Phys. B234189–212

  28. [36]

    Georgi H 1984Weak Interactions and Modern Particle Theory(Redwood City, CA: Benjamin/Cummings)

  29. [37]

    Manohar A V 1998 (Preprinthep-ph/9802419)

    1998

  30. [38]

    Aoki Y, T A, Bennett E, Fukaya H, Hashimoto Set al.2017Phys. Rev. D96014508

  31. [39]

    Weinberg S 1979Physica A96327–340

  32. [40]

    Shifman M A, Vainshtein A I and Zakharov V I 1978Phys. Lett. B78443–446

  33. [41]

    Hoferichter M, Klos, Men´ endez J and Schwenk A 2017Physical Review Letters119ISSN 1079- 7114

  34. [42]

    Aalbers Jet al.(LZ) 2023Phys. Rev. Lett.131041002

  35. [43]

    Aalbers Jet al.(DAR WIN) 2016JCAP11017

  36. [44]

    Sirunyan A Met al.(CMS) 2020JHEP05033

  37. [45]

    J.83844 (Preprint1611.03184)

    Ackermann Met al.(Fermi-LAT, DES) 2017Astrophys. J.83844 (Preprint1611.03184)

    arXiv

  38. [46]

    Acharyya Aet al.2021Journal of Cosmology and Astroparticle Physics2021057

  39. [47]

    Banks T and Zaks A 1982Nucl. Phys. B196189–204

  40. [48]

    Ryttov T A and Sannino F 2008Phys. Rev. D78065001

  41. [49]

    Hasenfratz A, Neil E, Shamir Y, Svetitsky B and Witzel O 2023Phys. Rev. D108L071503 (Preprint2306.07236)

    arXiv

  42. [50]

    Brower R Cet al.(Lattice Strong Dynamics) 2024Phys. Rev. D110054501 (Preprint 2306.06095)

    arXiv

  43. [51]

    Amaro-Seoane Pet al.(LISA) 2017arXiv:1702.00786(Preprinthttps://arxiv.org/abs/1702. 00786)

    arXiv

  44. [52]

    Witzel O (Lattice Strong Dynamics Collaboration) 2025PoSLA TTICE2024146 (Preprint 2412.10322)

    arXiv

  45. [53]

    Rev.175(5) 2195–2199

    Gell-Mann M, Oakes R J and Renner B 1968Phys. Rev.175(5) 2195–2199

  46. [54]

    Di Valentino Eet al.2021Class. Quant. Grav.38153001 (Preprint2103.01183)

    arXiv

  47. [55]

    Abdalla E o 2022Journal of High Energy Astrophysics3449–211 (Preprint2203.06142)

    arXiv

  48. [56]

    Yamanaka N, Iida H, Nakamura A and Wakayama M 2021Phys. Lett. B813136056 (Preprint 1910.01440)

    1910 arXiv

  49. [57]

    Yamanaka N, Iida H, Nakamura A and Wakayama M 2020Phys. Rev. D102054507 (Preprint 1910.07756)

    1910 arXiv

  50. [58]

    Arcadi Get al.2018The European Physical Journal C78ISSN 1434-6052

  51. [59]

    Gasser J and Leutwyler H 1984Annals of Physics158142–210 ISSN 0003-4916

  52. [60]

    Rev., 183: 1245-60(July 25, 1969)

    Dashen R 1968Phys. Rev., 183: 1245-60(July 25, 1969)

    1969

  53. [61]

    Hinterbichler K, Khoury J, Levy A and Matas A 2011Physical Review D84ISSN 1550-2368

  54. [62]

    Silveira V and Zee A 1985Phys. Lett. B161136–140

  55. [63]

    Peskin M and Schroeder D 1995An Introduction to Quantum Field TheoryAdvanced book classics (Avalon Publishing) ISBN 9780201503975 Naturally resonant two-mediator SIDM model with decoupled relic abundance42

  56. [64]

    Sorella S 1991International Journal of Modern Physics B05937–976

  57. [65]

    Hufnagel M, Schmidt-Hoberg K and Wild S 2018JCAP1811032 (Preprint1808.09324)

    arXiv

  58. [66]

    Winkler M W 2019Phys. Rev. D99015018 (Preprint1809.01876)

    arXiv

  59. [67]

    Ellis J R, Ferstl A and Olive K A 2000Phys. Lett. B481304–314 (Preprinthep-ph/0001005)

  60. [68]

    Griest K and Seckel D 1991Phys. Rev. D433191–3203

  61. [69]

    Paolo G and Graciela G 1991Nuclear Physics B360145–179 ISSN 0550-3213

  62. [70]

    Caswell W E 1974Phys. Rev. Lett.33244–246

  63. [71]

    Jones D R T 1974Nucl. Phys. B75531–538

Pith tools

Reviewed August 6, 2026 · model on record in the stance chip above.