claims depot shelf
Dark Energy
Formal claims (Lean)
-
Within the Wilsonian flow with the coarse-graining scale $\kappa$ identified with $H$, the paper claims that a flat-potential de Sitter equilibrium is a saddle point of the two-dimensional dynamical system $(h,\varepsilon)$, with eigenvalues $\lambda_{\mathrm{dS},\pm}=(-1\pm\sqrt{5})/2+O(\gamma)$ and a Minkowski scaling attractor $H\propto e^{-N/2}$; a universe that starts at rest in de Sitter is
-
The paper's central claim is that the unimodular Brans-Dicke model with the Coleman-Weinberg potential is a viable and observationally consistent inflationary model. For a spatially flat FRW universe in the unimodular frame the field equations admit power-law solutions $a/a_0 = (\tau/\tau_0)^{1/3}$ and $\varphi/\varphi_0 = (\tau/\tau_0)^{\xi/3}$ with $\xi = 1/(1+\gamma)$ during the $w=-1$ phase, a
-
On the paper's own terms, the central discovery is a negative result turned positive: an early-universe modification cannot remove the DESI late-time signal. In the full-data fit of the EDE-CPL model (CMB distance priors, DESI DR2 BAO, Pantheon+ supernovae with the SH0ES subsample), the CPL parameters are constrained to $w_0 = -0.85 \pm 0.06$ and $w_a = -0.56 \pm 0.23$, a clear departure from the
-
On the paper's own terms, the central discovery is a separation between statistical fit and dynamical behavior. All four emergent models, whose dark-energy or curvature terms switch on through a hyperbolic-tangent-type function, fit the low-redshift expansion data well enough to remain viable alternatives to $\Omega_k$-\Lambda CDM, with PEDE named as the best candidate by AIC and BIC in the abstra
-
The paper's central claim is that the cosmographic reconstruction of the quintessence potential, previously available only for minimal coupling, extends exactly to scalar-tensor gravity. In the Jordan frame with action $S=\int d^4x\sqrt{-g}\,\big[\tfrac12 F(\Phi)R - \tfrac12 \partial_\mu\Phi\partial^\mu\Phi - U(\Phi)\big] + S_m$, the four quantities $\lambda_U$, $\lambda_F$, $\Gamma_U$, and $\Gamm
-
The paper's central claim is that, for the exponential potential, the thawing equation of state is, through $O(\lambda^4)$, $w_\phi = -1 + f_2(\Omega_\phi)\lambda^2 + [f_2'(\Omega_\phi)\Omega^{(2)} + f_4(\Omega_\phi)]\lambda^4 + O(\lambda^6)$, where $\Omega^{(2)}$ is the $\lambda^2$ correction to the background density parameter and $f_4$ is a new closed-form function. Evaluating $f_2$ on the $\La
-
The central claim is that a single scalar field on a hyperbolic α-attractor plateau can mimic the premature termination of hybrid inflation by inserting a waterfall step in the potential, V_wf = V_original (1 + γ tanh((φ−φc)/Δφ))/(1+γ). Because the step is written in the half-plane variable T = exp(−√(2/(3α)) φ), it respects the hyperbolic geometry of α-attractors. The premature stop at φc removes
-
The paper claims that the divergence of the non-interacting dust collapse, $\rho_m \sim \chi^{-2}$, and the residual central mass term of the constant-interaction model are both eliminated by the radial interaction $\epsilon(\chi)=2(1+w)/\chi-\alpha(t)$. With this choice, the dust density at the center is $\rho_m(t,0)=-(1+w)\rho_{\rm de}^0(t)$, the total density is $\rho_t(t,0)=-w\rho_{\rm de}^0(t
-
The central discovery is that vacuum-energy cancellation is a property of the projectable-lapse structure and the global constraint it generates, not of ultra-locality itself. Adding the leading normal-derivative operators to the gravitational and flux sectors, the paper derives effective Einstein equations in which the matter effective action enters only through its dynamical part, with the vacuu
-
The paper establishes that the generalized second law, applied to a general $f(H)$ modified Friedmann cosmology, is not merely a consistency check but a selection criterion for horizon entropy. The non-equilibrium GSL is rewritten in terms of the Hubble parameter, the deceleration parameter, and the total equation-of-state parameter $\omega$, and then mapped to the apparent-horizon area $A$. Assum
-
The paper establishes that the generalized second law, applied to a general $f(H)$ modified Friedmann cosmology, is not merely a consistency check but a selection criterion for horizon entropy. The non-equilibrium GSL is rewritten in terms of the Hubble parameter, the deceleration parameter, and the total equation-of-state parameter $\omega$, and then mapped to the apparent-horizon area $A$. Assum
-
On the paper's own terms, the central discovery is that a nonrelativistic test particle moving in the gravitational field of a point mass immersed in the quantum spin-connection foam obeys the modified law $a = \sqrt{(GM/r^2)^2 + \bar a^2}$, where $\bar a^2 = \langle \tilde\omega^2\rangle$ is the variance of the fluctuating spin-connection components. Re-expressing accelerations relative to the me
-
The central claim is that viability of fractional holographic dark energy depends sharply on which infrared cutoff defines the horizon length in the density formula. With only low-redshift data (SN+OHD+DESI DR2), all three variants—Hubble-horizon FHDEH, future-event-horizon FHDEF, and particle-horizon FHDEP—fit about as well as ΛCDM, each giving a marginally lower χ²_min. Adding Planck CMB distanc
-
The central claim is that, in the coincident-gauge flat FLRW branch of the exponential $f(Q)$ model with $n=1$, apparent-horizon dynamics admits an equilibrium thermodynamic description whose entropy is not the bare area law. Projecting Hayward's unified first law along the horizon tangent with the effective Misner--Sharp--Hernandez mass $M_{\rm MSH}^{(\rm eff)}=R^3(Qf_Q-f/2)/6$ gives an entropy d
-
The central claim is that, in the coincident-gauge flat FLRW branch of the exponential $f(Q)$ model with $n=1$, apparent-horizon dynamics admits an equilibrium thermodynamic description whose entropy is not the bare area law. Projecting Hayward's unified first law along the horizon tangent with the effective Misner--Sharp--Hernandez mass $M_{\rm MSH}^{(\rm eff)}=R^3(Qf_Q-f/2)/6$ gives an entropy d
-
GCT modifies gravity from the fluid side by coupling a vector field $S_\mu$ to the standard energy-momentum tensor through a source term proportional to $(\partial w_f)^2$, where $w_f$ is the total fluid equation of state. In the radiation- and matter-dominated eras this source vanishes and the field decays to a stable critical point, while during the radiation-matter transition it drives a transi
-
The central claim is that the dark energy driving late-time acceleration can be an emergent geometric component rather than a cosmological constant or a phenomenological parametrization. In the CKG framework the divergence-free conformal Killing tensor of the Robertson–Walker spacetime behaves as a perfect fluid with density $\mu_D = -\frac{1}{2} C a^2 + \Lambda$ and pressure $p_D = \frac{5}{6} C
-
On the fixed RN–AdS–Kiselev geometry, with a negatively charged test particle in a monotone electrostatic potential $\Psi(r)$, the paper establishes three claims. The generalized electrostatic ergosphere defined by $|q|\Psi(r_E)=\sqrt{f(r_E)}$ has a unique exterior boundary, and this boundary moves outward with increasing charge magnitude, mass, black-hole charge, quintessence amplitude, and more
-
The paper's central claim is that isotropic cosmic birefringence, $\beta\approx0.3^\circ$, can be produced by a 3-form dark-energy field through the parity-violating interaction $\mathcal{L}_{\mathrm{int}}=\epsilon\lambda\,C_{\mu\nu\rho}F^{\mu\nu}A^{\rho}$. In a homogeneous isotropic background the 3-form reduces to a single function $\chi(t)$, with $C_{ijk}=a^3\chi\,\epsilon_{ijk}$, and the inter
-
The central claim is that a two-axion dark sector with a light dark-energy axion and a heavy dark-matter axion can be recast exactly as coupled quintessence, provided the heavy axion stays on its WKB branch. The interaction endows the dark-matter mass with a phi dependence given by $m_\chi(\phi) = m_\chi \sqrt{1 + \beta \cos(\phi/f_\phi)}$, Eq. (11), so the dark-matter density follows $\rho_\chi/\
-
On the paper's own terms, the discovery is that for all five dark-energy parameterizations—JBP, FSLL I, FSLL II, polynomial (SL), and BA—with parameters fixed by the Constitution, Union 2, and LOSS supernova fits, the perturbation observables behave in a common, distinctive way. The growth index varies slowly with redshift, with $\gamma(0)$ between about 0.537 and 0.547 (close to $6/11$) and $\gam
-
The central discovery is a joint dark-sector degeneracy: late-time BAO and supernova distances prefer $w_{\mathrm{DE}}>-1$, while the CMB fixes the early-Universe matter density around last scattering; a positive $w_{\mathrm{DM}}$ changes how that early density maps to the present, lowering the dark-matter density at $z=2.33$ and increasing the matter-era distance interval by about 0.2%, which rel
-
The central discovery is a joint dark-sector degeneracy: late-time BAO and supernova distances prefer $w_{\mathrm{DE}}>-1$, while the CMB fixes the early-Universe matter density around last scattering; a positive $w_{\mathrm{DM}}$ changes how that early density maps to the present, lowering the dark-matter density at $z=2.33$ and increasing the matter-era distance interval by about 0.2%, which rel
-
The central discovery is a structural degeneracy between intrinsic black hole spin and the surrounding dark matter pressure. The paper constructs a piecewise three-region spacetime: an inner vacuum up to r_b, an intermediate shell of Modified Chaplygin Gas (P = Aρ − B/ρ^n) with the mass and pressure profile obtained from a coupled TOV integration, and an outer vacuum matched at the radius r_s wher
-
On its own terms, the paper's central claim is that the analytically tractable Extended Cuscuton sector provides a controlled, minimally modified dark-energy target that remains close to $\Lambda$CDM but is not observationally inert. The four benchmark submodels defined by the sign of $\tilde c_1$ and by the two asymptotic de Sitter branches all fit the combined CC+BAO+SN data with $\Omega_\Lambda
-
The central discovery is that a specific interacting dark-sector theory—a quintessence field coupled to dark matter by the conformal transformation g̃μν = e^{-2αφ}gμν, with a potential engineered so ρφ = Λ + ρDM - ρc—can simultaneously keep the ΛCDM background and lower late-time structure growth enough to reconcile early-universe CMB measurements with growth data. In the fit, the conformal coupli
-
The paper's central discovery is a replication: taking the numbers (264°, 48°) from the criticized analysis and treating them as equatorial right ascension and declination exactly reproduces the reported hemisphere split of 724/840 supernovae in the N=1564 Pantheon+ sample, whereas the CMB dipole direction properly converted from Galactic to ICRS coordinates, (167.8°, -7.1°), yields 539/1025. This
-
The central assertion is that an exactly FLRW background can coexist with a matter sector that has a preferred spatial direction, because isotropy is realized only on shell. For a vector-field Lagrangian of the form (2.4), the conditions (2.8) cancel the anisotropic stress for arbitrary homogeneous configurations, and the temporal vector equations (together with an off-shell identity from Ref. [32
-
The paper establishes that a constant 'radioactive' decay rate Γ for dark energy, measured in units of the Hubble rate H0, is compatible with the full current data set. For distance data alone (DESI DR2 BAO plus supernovae), the posterior for Γ/H0 shifts positive by about 2σ, corresponding to a dark-energy density that decreases with time and an effective equation of state w > -1 at low redshift.
-
The paper's central claim is that the density-level pivot construction applied to CPL removes the parameter correlation that made the standard $(\omega_0,\omega_a)$ basis look degenerate, and that in this physically motivated basis $\omega_0\omega_a$CDM remains favored over $f_af_b$CDM and reproduces quintessence backgrounds more accurately. The transformation is exact: $(\omega_0,\omega_a)\leftri
-
On the paper's own terms, the discovery is that after subtracting a redshift-dependent magnitude offset of the form Δμ = 0.183(1 − e^{−2.2z}) from each distance modulus, the corrected Pantheon+ and DES-SN5YR Hubble diagrams are described almost exactly by the closed form d_L = (c/H0)(1+z)ln(1+z). Fitting only H0 per survey gives lower χ² than a two-parameter flat ΛCDM model; an extended quadratic
-
In the joint CMB-SPA + DESI BAO + full-shape analysis, allowing free spatial curvature yields Ωk = (3.0 ± 1.1) × 10^{-3} and ns = 0.9692 ± 0.0035. This value is 1.8σ lower than the flat-ΛCDM result and brings Starobinsky and Higgs inflation back inside 2.3σ and 1.5σ, respectively. The same data therefore favor a slightly open universe at 2.7σ, and the apparent conflict with plateau inflation is an
-
For the representative KGB model they simulate, braiding enhances dark-energy clustering, raises the Weyl-potential amplitude, and slows its time evolution. That produces up to roughly 12% more weak-lensing convergence power than k-essence at multipoles around 100–1000, while the ISW–RS signal is suppressed by tens of percent in the linear regime and then overtakes k-essence once nonlinear evoluti
-
Within the Swampland framework, the paper's central claim is that invoking an ultralight axion to explain the suggested CMB polarization rotation ϑ ∼ 10^-3 radians is in genuine tension with the magnetic Weak Gravity Conjecture bound Λ³_QCD ≲ m f M_Pl. For decay constants f ∼ 10^11–10^16 GeV this bound forces m ≳ 10^-28–10^-23 eV, right next to the mass range m ≲ 10^-28 eV that the birefringence s
-
Once reconstructed with f(Q,T)=αQ^m+βT and a hybrid scale factor, the model produces a nonsingular asymmetric bounce near t≃−0.09, where the Hubble parameter flips from negative to positive, then evolves so the effective equation of state approaches −1 at the present age, while energy conditions behave as required for a bounce followed by late acceleration.
-
Starting from the luminal Horndeski action with G2 = a1 X + a2 X^2 - V(phi), G3 = 3 a3 X box phi, and a constant Planck mass, plus the covariant interaction beta Z^2 with Z = u_c^mu grad_mu phi, the paper shows that the background evolves from w_DE ~ 1/6 in the radiation era to a stable phantom phase with w_DE < -1 at intermediate redshifts, then returns to w_DE > -1 at z_c ~ 0.36-0.66 as the expo
-
The central claim is that in the ΛXCDM model the effective equation of state of the composite dark-energy fluid, w_eff = −1 + (1+w_X) Ω_X/Ω_D, necessarily crosses the phantom divide from phantom-like to quintessence-like behavior as Ω_X changes sign, provided the cosmon X behaves as phantom matter (w_X < −1, Ω_X < 0 today) and the vacuum runs with coefficient ν < 0 (equivalently ϵ = ν(1+w_X) > 0).
-
The authors integrate the GREA background directly into a Boltzmann solver and evolve the entropic component as an effective dark-energy fluid with sound speed c_s² = 1, regulated by the parametrized-post-Friedmann scheme so perturbations remain regular through the phantom crossing. The resulting angular power spectra and growth functions are new. When fit to the CMB-SPA (Planck+ACT+SPT) likelihoo
-
On its own terms, the paper establishes that the GEDE extension is not favored by current data. For the CMB+DESI DR2 combination alone, the preferred Δ is 0.46±0.25, about 1.8σ from ΛCDM; adding any of the three supernova samples pulls Δ to within 0.06–0.33σ of zero. The Hubble constant comes out at 67.9–69.8 km/s/Mpc, the sound horizon at 147.4–147.6 Mpc, and S8 at about 0.821, all consistent wit
-
The central claim is that the entropy-area relation with a scale-dependent exponent Δ(x)=Δ0+Δ1/ln x, where x=(H1/H)^2, when fed into the gravity-thermodynamics first law on the apparent horizon, yields an exact analytic Hubble rate H(z). Combined fits to Hubble-rate measurements, type Ia supernova distances, baryon acoustic oscillations, and CMB shift parameters give H0 = 69.27^+0.62_-0.65 km/s/Mp
-
The central discovery is that a cosmological model in which the Gauss-Bonnet correction to the Wald entropy of the apparent horizon is sourced by the astrophysical history of black-hole formation and mergers fits the combined early- and late-Universe data better than ΛCDM, with the coupling parameter Cn = 0.435^{+0.150}_{-0.132}, a ~3σ phantom-side deviation from zero. The mechanism preserves the
-
The central discovery is that a cosmological model in which the Gauss-Bonnet correction to the Wald entropy of the apparent horizon is sourced by the astrophysical history of black-hole formation and mergers fits the combined early- and late-Universe data better than ΛCDM, with the coupling parameter Cn = 0.435^{+0.150}_{-0.132}, a ~3σ phantom-side deviation from zero. The mechanism preserves the
-
On the kinematic branch HL = 1, for any barotropic background with w_b > -1 and any entropy modification whose local logarithmic slope satisfies chi_cr < 2, the phantom-divide crossing is unique inside the physical phase space 0 < Omega_X < 1, occurs at a strict local maximum of the event-horizon radius, proceeds from quintessence into phantom, and yields the exact local entropy-scaling dimension
-
On the kinematic branch HL = 1, for any barotropic background with w_b > -1 and any entropy modification whose local logarithmic slope satisfies chi_cr < 2, the phantom-divide crossing is unique inside the physical phase space 0 < Omega_X < 1, occurs at a strict local maximum of the event-horizon radius, proceeds from quintessence into phantom, and yields the exact local entropy-scaling dimension
-
By consolidating the normal and self-accelerating DGP branches, cubic scalar and vector Galileons, the generalised cubic covariant Galileon and generic EFT parameterisations into a single master Vainshtein equation whose physics is carried by three time-dependent background functions, EFT-RAMSES produces nonlinear matter power spectra that agree with both its parent code and an independent approxi
-
By consolidating the normal and self-accelerating DGP branches, cubic scalar and vector Galileons, the generalised cubic covariant Galileon and generic EFT parameterisations into a single master Vainshtein equation whose physics is carried by three time-dependent background functions, EFT-RAMSES produces nonlinear matter power spectra that agree with both its parent code and an independent approxi
-
The central claim is that if the bare vacuum energy is negative and sampled uniformly, and if universes are weighted by cumulative observation time, then the axion mass becomes anthropically confined to roughly one decade around the Hubble scale, and the probability of observing 0.1 < Ω_m < 0.9 is about 40%. The same logic anthropically disfavors slow-roll dark energy (which needs fine-tuned initi
-
The central claim is that if the bare vacuum energy is negative and sampled uniformly, and if universes are weighted by cumulative observation time, then the axion mass becomes anthropically confined to roughly one decade around the Hubble scale, and the probability of observing 0.1 < Ω_m < 0.9 is about 40%. The same logic anthropically disfavors slow-roll dark energy (which needs fine-tuned initi
-
When interacting ghost dark energy evolves inside the Kaniadakis-corrected flat Friedmann equations, the deformation parameter λ produces only mild shifts in the equation-of-state and deceleration histories, moderates the classical instability measured by the squared sound speed, and drives the statefinder trajectory toward the ΛCDM point {r,s}={1,0} with smaller present-day deviations as λ grows.
-
A new class of Morris–Thorne wormholes sourced by fractional holographic dark energy, with redshift φ(r)=−0.1/r, satisfies the throat, flare-out and asymptotic-flatness conditions, remains free of curvature singularities, and becomes progressively less exotic and more thermodynamically regular as the fractional parameter γ is increased.
Stated claims
-
The paper discovers that any canonical scalar field coupled to dark matter through its mass, which produces an apparent phantom crossing (w_eff < -1) in the effective dark energy, must have a minimum coupling strength beta_min that depends only on the observed expansion history H(z) and is independent of the potential V(phi). This is derived from the identity phi'^2/3 = B(z) + Omega_c_tilde(z) $\Delta$(z) (Eq. 10), where B(z) is determined solely by H(z) (Eq. 12). The non-negativity of kinetic energy forces $\Delta$(z) >= -B(z)/Omega_c_tilde(z) (Eq. 13), giving a minimum mass loss. Then, because the mass changes only when the field moves, and the field speed is itself set by H(z) via K = B + Omega_c_tilde $\Delta$, the mass loss obeys a differential equation (Eq. 17) with a speed limit proportional to |$\beta$|. This yields a minimum coupling beta_min[H(z)] (the smallest |$\beta$| for which Delta_max >= Delta_min at all redshifts). For DESI DR2, this gives beta_min ~ 0.05, and the bound is sharp: it is saturated when the field comes to rest at the depth of the crossing (z ~ 0.8).
-
The paper establishes that a vacuum–dark-matter interaction, with the energy transfer unperturbed in the CDM rest frame (δQ = 0), provides a mechanism to lower S8 independently of the early expansion history. When the vacuum decays into CDM, the newly created dark matter carries no density perturbation, diluting the existing contrast: Ωm increases while σ8 decreases, and the net S8 falls. For the combined data set CBS (Planck+BAO+Pantheon+) the posteriors favour m̄ < 0 at 1.8–2.7σ, and the exchange lowers S8 by 0.02–0.08 without raising H0. With the SH0ES calibration (CBSH), the exchange removes the S8 increase of dark radiation (S8 = 0.800 ± 0.016 at H0 = 71.34 ± 0.70 km/s/Mpc) and about a third of that caused by early dark energy (for α = −2). However, the S8 reduction is largest today and amounts to only 1–2% at the redshifts where the KiDS-Legacy lensing kernels peak, and the exchange does not ease the Hubble tension itself.
-
The central discovery is that galaxy clusters, as highly biased tracers of large-scale structure, exhibit a clear BAO signal in their two-point correlation function at z=0.25 and 0.43, allowing isotropic distance measurements with uncertainties of 7–9%. The detected BAO scale matches the standard ΛCDM prediction, and the joint analysis with other probes yields dark energy parameters (w0 = –0.883 ± 0.091, wa = –0.48+0.47–0.41) that are consistent with a cosmological constant. This establishes the utility of clusters for BAO cosmology and provides the largest such dataset to date.
-
The paper establishes that if dark energy can have spatial variations of order 10–20% on scales of hundreds of megaparsecs, then the local expansion rate of a void becomes measurably different from the prediction of the homogeneous dark energy model even when the matter density profile is identical. This difference arises because a dark energy deficient region has a smaller local expansion rate than a void with the same matter content evolving in a homogeneous dark energy background. The key relation is H²_v – H²_{vhom} = –H² Ω_DE A_DE W_v, where A_DE parametrizes the deficiency. For a model void with central matter contrast δ_c ≈ –0.73, a 10% deficiency reduces the expansion rate by ~2.1 km/s/Mpc (a 2.7% suppression), producing a radial velocity shift of ~62 km/s at 30 Mpc. Deficiencies of ~20% extending over 600 Mpc could shift the locally inferred Hubble constant from 67.4 to 73.2 km/s/Mpc, but only if the observer is sufficiently centered to avoid generating excessive directional anisotropy in the Hubble diagram and CMB.
-
The central claim is that cosmic acceleration is transient in a spatially flat universe whose dark energy follows a specific oscillatory equation of state. The equation of state $w_{\rm de}(N)$ is the solution of $\frac{d^2w_{\rm de}}{dN^2} + f^2(w_{\rm de} - w_{\rm m}) = \Omega_{m0}N^3$ with mean $w_{\rm m} = -1$, where $N = \ln a$ counts e-folds and the source term is normalized to the present matter density parameter. The source term is negligible near the present but dominates at large $|N|$: toward the past it drives the oscillations into the phantom regime $w_{\rm de} < -1$, and toward the future it pushes them into $w_{\rm de} > -1$. Once $w_{\rm de}$ rises above $-1$, the continuity equation forces $\rho_{\rm de}$ to dilute, $H$ decreases asymptotically to zero, and the expansion decelerates permanently, approaching a static state as $a \to \infty$ without contraction. Extrapolated to the future, the BellDE and CPL parametrizations studied alongside show the same replacement of $\Lambda$CDM's eternal de Sitter future, with the sourced-oscillation model extending the accelerating phase by a few e-folds. With Planck and DESI BAO plus one supernova compilation at a time, the model constrains $H_0$ to about $66.1$–$67.8\ \mathrm{km\,s^{-1}\,Mpc^{-1}}$ and $\Omega_m$ to about $0.31$, consistent with the baseline model; all dynamical models sit at lower $H_0$ than $\Lambda$CDM, so none of them relieves the Hubble tension.
-
The paper claims that the tachyonic DBI dark energy scenario — a single scalar field with Lagrangian $\mathcal{L}_\phi = -V(\phi)\sqrt{1-2X}$, equation of state $w_\phi = \dot\phi^2 - 1 \in (-1,0)$, and sound speed $c_s^2 = 1 - \dot\phi^2 < 1$ — is observationally viable against five complementary probes, with all standard cosmological parameters within $2\sigma$ of $\Lambda$CDM (the largest shift being $H_0 \simeq 67.4$ versus $68.6\ \mathrm{km\,s^{-1}\,Mpc^{-1}}$ for Pantheon+, about $1.8\sigma$) and with Bayesian evidence favoring it over $\Lambda$CDM at $\Delta\ln\mathcal{Z} \simeq +3.38$ for Pantheon+ (moderate), and $\simeq +0.64$ and $+0.66$ for DES-SN5YR and Union3 (inconclusive). Dark energy perturbations genuinely grow, reaching $|\delta_{\rm de}|/|\delta_m| \sim 10^{-6}$ at $z=0$, but their Poisson source term $\Omega_{\rm de}|\delta_{\rm de}|$ stays four to six orders of magnitude below the matter source, so the subluminal sound speed acts only through a subtle modification of the Hubble friction in the matter growth equation, shifting $S_8$ upward by $\approx 0.5\sigma$ and leaving the $S_8$ tension unrelieved. Four diagnostics — a CPL projection with negative $w_a^{\rm eff}$, a negative $Om(z)$ slope, statefinder coordinates off the $\Lambda$CDM fixed point $(1,0)$, and a growth index $\gamma_0 \approx 0.555$–$0.556$ above the general-relativistic $6/11$ — are presented as independent evidence that the field is genuinely thawing quintessence rather than a disguised cosmological constant. The paper acknowledges that $w_\phi \ge -1$ bars phantom crossing, consistent with the single-field no-go theorem, so the model cannot follow the DESI DR2 preference for an effective equation of state crossing $w=-1$.
-
The central claim is that modified symmetric teleparallel gravity—where gravity is described by non-metricity rather than curvature—can reproduce the observed late-time accelerated expansion of the universe without a cosmological constant. By choosing specific functional forms of f(Q) (logarithmic, polynomial, tanh-power) and scalar-tensor extensions f(Q,φ), the thesis demonstrates through dynamical systems analysis that the models contain stable critical points corresponding to an accelerating universe with ω_eff = -1. The Noether symmetry approach further yields exact cosmological solutions that describe accelerated expansion, and the computed Hubble parameter, deceleration parameter, and effective equation of state are shown to align with current observational constraints.
-
The paper discovers that for the dissipation coefficient scaling as $\Gamma_\xi \propto T^{-1}$, the two-scalar-field warm inflation model yields $(n_s,r)$ values within the 68% and 95% confidence regions of the combined Planck, ACT, DESI, and BICEP/Keck data, while for $p=1$ and $p=3$ the predictions lie outside the observationally allowed region. Compared to standard single-field warm inflation, the two-field model produces a systematically larger spectral index $n_s$ at fixed tensor-to-scalar ratio $r$. The viable scenarios cover a range of dissipation ratios $Q_{\xi*} \lesssim 6.29$, with $r$ between $2\times10^{-5}$ and $2\times10^{-3}$, within reach of next-generation CMB polarization experiments.
-
On its own terms, the paper establishes that a single separable Horndeski action — $L=\tfrac12 R + V_k(\varphi)X^n - V_g(\varphi)X^m\square\varphi$ with exponential $V_k, V_g$ and exponents linked by $m=(2n-1)/6$ — can reproduce the $w_0w_a$CDM phenomenology while removing the need for a cosmological constant or scalar potential. The central mechanism is a scalar-field charge density $C=\dot\varphi K_X + 6Hg - 2\dot\varphi G_{3\varphi}$ that dilutes as $a^{-3}$ in the early universe, so the theory tracks its shift-symmetric limit at early times and only breaks symmetry late, when the source term $P_\varphi$ grows. Analytic fixed-point analysis gives the radiation- and matter-era values of $w$, shows that histories converge to the $f=2n$ ray at matter domination so the braiding fraction $f$ is not an effective parameter, and identifies the de Sitter end state; the phantom-crossing condition $2nx_k = x_g(x_\varphi - 1 + u/3m)$ requires $u$ and $r$ to have opposite signs with $r\le0$. With only two dark-energy parameters ($r$ and $u_{\rm ref}$) — the same number as $w_0w_a$CDM — the best fits cross $w=-1$ at $z\approx0.5$ and improve the fit by $\Delta\chi^2\approx-3$, with modest, testable deviations in the gravitational coupling $G_{\rm eff}$ from general relativity.
-
The paper's claim is that growth-rate measurements distinguish dark energy from matter even when the two are degenerate at the background level, because the k-essence scalar field clusters differently from cold dark matter. Working in the quasi-static subhorizon regime, the growth equation keeps its standard form with the k-essence sector entering only through $H(z)$, and the observable $f\sigma_8(z)$ is fit to RSD data with stability conditions (no ghosts, no gradient instabilities) imposed as theory priors inside the MCMC. The result is that BAO measurements largely break the degeneracy in the ghost condensate (Model 1), while RSD measurements are crucial for the dilatonic ghost condensate (Model 2), where adding RSD to CC+PN+BAO reduces the width of the $\Omega_{m0}$ 68% credible interval by approximately a factor of 2.2. The paper also reports that including BAO pushes Model 1 toward its $\Lambda$CDM limit while shifting Model 2's parameter $\lambda$ away from $\Lambda$CDM, with a $\Delta\chi^2 \simeq -4.1$ improvement over $\Lambda$CDM that the BIC does not reward given the extra parameters.
-
The central discovery is the first successful implementation of Horndeski gravity in a cosmological N-body code via the effective fluid approach. By rewriting the linear dark energy perturbations ($\delta\rho_{\rm DE}$, $\delta P_{\rm DE}$, $\theta_{\rm DE}$, and $\sigma_{\rm DE}$) in terms of metric perturbations in the synchronous gauge, the code evolves these perturbations using the Boltzmann equation. The simulations show that for $\Lambda$CDM, the matter power spectrum agrees with standard General Relativity to within 0.2%, while for $f(R)$ Hu-Sawicki and cubic Galileon models, deviations reach up to 40% and 25% respectively, consistent with linear theory predictions.
-
For the general class of Lagrangians $F(b,Y,X)$ describing a perfect fluid interacting with a superfluid via derivative couplings, stable cosmological evolution requires a positive adiabatic sound speed $c_a^2 > 0$. During slow-roll expansion ($\epsilon_1, |\epsilon_2| \ll 1$), the relation $c_a^2 = -1 + \frac{2}{3}\epsilon_1 - \frac{1}{3}\epsilon_2$ forces $c_a^2 \simeq -1$, violating stability. Thus, slow-roll acceleration (whether for inflation or late-time dark energy) is ruled out. Moreover, the stability conditions imply a stronger null energy condition $\rho+P > 2Y^2F_X > 0$, which forbids the total medium from crossing $w=-1$, and when the medium is split into a pressureless CDM component and an effective dark energy component, the transition can only occur from $w_{\rm DE}>-1$ to $w_{\rm DE}<-1$, contradicting the crossing direction hinted by DESI analyses.
-
Stated on the paper's own terms, the central claim is that the integrated three-point correlation function $\zeta_\pm$ of cosmic shear can be analysed on survey data with a fully simulation-based approach, and that it adds genuine cosmological information on top of the established two-point analysis. The evidence is the first survey detection of $\zeta_\pm$, at signal-to-noise 12.8, from the DES Y3 shape catalogue, produced by a blinded pipeline whose forward model includes the survey mask, shape noise, intrinsic alignments, baryonic feedback, photometric redshift uncertainties and shear calibration bias. Adding $\zeta_\pm$ to $\xi_\pm$ tightens the $\Omega_m$–$S_8$–$w_0$ constraints and improves the figure of merit by a factor of 1.59, as predicted by the simulated analyses of the companion paper. The gain is delivered on large angular scales — at least 15 arcmin separations and aperture radii of at least 50 arcmin — where baryonic effects are subdominant, so the improvement does not rely on small, baryon-sensitive scales. The resulting posteriors are consistent with other DES Y3 analyses of the same source catalogue and with KiDS-1000, and differ from Planck 2018 + BAO by at most $2.4\sigma$ in the three-dimensional $\Omega_m$–$S_8$–$w_0$ space.
-
The paper's claim is that a second-order derivative expansion of the likelihood is accurate enough to replace full sampling in bright-siren cosmology, and dramatically more accurate than the Fisher matrix. Working with 300 events per catalog out to $z=1$ and a Cosmic Explorer plus triangular Einstein Telescope network, the authors report an $H_0$ precision of about 2.7 percent with the Doublet against about 2.5 percent for the exact likelihood, and they quote that the Doublet cosmological uncertainties differ from the exact ones by only about 10 percent on average across the four cosmological parameters, compared with about 195 percent for the traditional Fisher approximation — roughly a factor of nineteen in average closeness. The failure is not uniform across parameters: the $H_0$ width is comparatively stable under all three treatments, whereas Fisher broadens $ω_0$ and $ω_a$ substantially, so the beyond-Fisher correction matters most for dark energy. The authors also report a consistent ordering in Jensen–Shannon divergence from the exact posterior, Doublet below Singlet below Fisher, preserved across all three ways of building the gravitational-wave catalog, which they read as evidence that the ordering is a property of the approximation scheme rather than of the data. They acknowledge that the Doublet distances are systematically displaced toward smaller luminosity distances, producing larger parameter biases than Fisher at the same time as smaller uncertainties, and they attribute this to the adopted priors on distance and inclination rather than to the truncation alone.
-
The central claim is that direct determinations of $H_0$ are model-dependent because the apparent magnitude of a standard candle is $m = M + 5\log_{10}(d_L/10\,\mathrm{pc})$, with the luminosity distance $d_L = (1+z)\int_0^z dx/H(x)$ carrying the entire expansion history $H(x)$ and not just the local expansion rate. Within $\Lambda$CDM, varying the one extra parameter $\Omega_M$ moves the inferred $H_0$ only slightly. But in $w_0w_a$CDM, where the dark-energy term depends on both $w_0$ and $w_a$, moving along the degeneracy direction of the DESI ellipse in the $w_0$-$w_a$ plane changes the inferred $H_0$ significantly. The authors emphasize that this claim is not disputed in the paper they are answering, and that their earlier work was only that the background cosmology *can* impact local $H_0$ determinations and *may* be relevant to the Hubble tension, not that it resolves it.
-
The central claim is that the axion-monopole interaction does not create a new stable accelerated endpoint: every finite-Hubble critical point of the autonomous system has vanishing axion kinetic energy and lies at an extremum of the axion potential, and only the minima can be stable attractors. Those minima are stable exactly when $\Omega_{\Lambda 0} > 0$ and are asymptotically driven by the cosmological constant alone, with the axion at rest at its potential minimum; the de Sitter point at the potential maximum is always a saddle even though it survives at $\Omega_{\Lambda 0} = 0$. With no cosmological constant, the eventual fate is a damped, oscillating axion that behaves like pressureless matter, so acceleration is transient. The paper further establishes that matter-era tracking is an attracting slow-drift solution for the axion velocity rather than an exact critical point, sustained by the monopole-induced force with energy flowing from monopoles to the axion, and that the interaction-effective equation of state $w_{\rm DE}^{\rm eff}$ can dip below $-1$ while the physical $w_{\rm DE}$ stays at or above $-1$ and $\dot H \le 0$. Repeating the analysis with the full BPS dyon mass preserves these features while making the interaction strength and field evolution depend on the dyon charge state.
-
The central claim is that the gevolution-style relativistic $N$-body framework can be generalized from its Horndeski subclasses to the general linear Horndeski $\alpha$-basis: the scalar degree of freedom is evolved with the linear Horndeski perturbation equation alongside fully nonlinear matter and metric fields, with hi_class supplying the background, the $\alpha$-functions, and the initial conditions. Working with luminal models ($\alpha_T=0$), the paper shows that a running effective Planck mass ($\alpha_M\neq 0$) generates gravitational slip at linear order through the scalar anisotropic stress $\Pi^{\phi}_{ij} = (M_*^2/a^2)\,D_{ij}[-\alpha_T\Phi + (\alpha_M-\alpha_T)H\pi]$, and that $\alpha_M$ enters the friction terms of the vector and tensor metric perturbations. It further shows that the braiding term $\alpha_B H\nabla^2\pi$ drives strong small-scale clustering of the effective dark-energy density, which traces the cosmic web in real space, and that the evolution of $M_*^2$ shifts the normalization and scale dependence of the dark-energy density power spectrum. Validation consists of matching hi_class in the linear regime and recovering the previously implemented nonlinear $k$-essence and KGB limits, across three phenomenological and three covariant models.
-
On the paper's own terms, the discovery is that the revised Medium Tier configuration preserves, and in the multiprobe version extends, kinematic lensing's advantage over traditional weak lensing. In ΛCDM, KL shrinks the marginalized errors on $\Omega_m$ and $\sigma_8$ by factors of 2.2 and 1.4 and raises $\mathrm{FoM}_{\Omega_m\sigma_8}$ from 22,200 (WL) to 49,857 (KL); in $w_0w_a$CDM, $\mathrm{FoM}_{w_0w_a}$ goes from 10.95 (WL) to 22.54 (KL). Using the KL sample as both lens and source sample in a 3×2pt analysis (cosmic shear, galaxy clustering, galaxy-galaxy lensing) yields $\mathrm{FoM}_{w_0w_a}=34.58$, and increasing from 10 to 20 tomographic bins raises it to 45.93 provided the clustering model goes beyond the Limber approximation. One year of Medium Tier data (700 square degrees) already gives KL-alone constraining power comparable to the full-survey WL forecast. The engine behind these numbers is a per-galaxy shape noise of $\sigma_{\rm KL}^{\epsilon}=0.05$ estimated from the roll-angle simulations, compared with 0.37 for WL, which moves KL into the cosmic-variance-dominated regime on most scales, with the consequence that baryonic feedback becomes the main limiting systematic.
-
On the paper's own terms, the central claim is that at low redshift ($z < 1.6$) a purely magnitude-limited sample, selected as $H < 21.8$ with a three-dimensional colour-colour star-galaxy separation, performs as well as the standard suite of DESI-like colour-selected tracers. Tuned to the same total target density as the combined BG, LRG, and ELG samples, MagMax recovers comparable precision on $f\sigma_8$ and on the BAO Alcock-Paczynski parameters, and yields roughly ten percent tighter constraints on the dark-energy equation of state ($\sigma(w_0)=0.0675$, $\sigma(w_a)=0.5$ when quasars are included), because its lower galaxy bias strengthens the redshift-space-distortion signal at low redshift and the measurements are cosmic-variance limited up to $z=1.6$. At high redshift, the three LBG dropout samples deliver an effective constraint of $\sigma(f_{\rm NL}^{\rm loc}) \sim 1$ and neutrino-mass precision that improves from 20 to 10 meV as survey time grows, with the two science cases responding differently to depth. The paper's stated main outcome is an optimisation framework that links instrumental capabilities and observing time directly to cosmological constraining power, applicable to the design of future Stage-V surveys.
-
The central claim is that a scalar field can couple to the teleparallel translational gauge potential through a field-dependent effective tetrad $h^\phi_{A\mu} = h_{A\mu} + \alpha(\phi) \mathcal{B}_{A\mu}$, where $\alpha(\phi)$ is an arbitrary dimensionless function and $\mathcal{B}_{A\mu}$ is the gauge-invariant combination of the potential and Stückelberg fields. This preserves local translational invariance and reduces to the universal coupling when $\alpha=0$. In flat FLRW cosmology, the non-universal coupling modifies the scalar-sector scale factor to $a_\phi = a + \alpha(\phi)(a-1)$. The Stückelberg equation provides a first integral $\alpha a_\phi^3 \rho_\phi = \text{constant}$, which governs the evolution of $\alpha$. In the early Universe, the coupling vanishes as $a\to 0$, recovering standard stiff-fluid behavior. In potential-dominated regimes, accelerated expansion requires the spatial deformation factor $m = a_\phi/a$ to satisfy $0<m<3$, and potential domination itself is controlled by the logarithmic slope of $\alpha$. The framework also admits solutions where departures from universality are dynamically suppressed, including power-law inflation and asymptotically de Sitter expansion. When a universally coupled perfect fluid is included, the non-universal scalar sector generates a matter-like contribution scaling as $a^{-3}$ while the background approaches de Sitter at late times.
-
On the paper's own terms, the discovery is that the screening charge of a nucleon, the factor $Q$ by which its scalar coupling is suppressed, is controlled by the unscreened mass beyond a screening radius, so a source with a long dilute tail behaves qualitatively differently from a uniform sphere. For the Skyrme trace profile $\rho_{\mathrm{tr}} = \rho_A$ inside $r < a$ and $\rho_{\mathrm{tr}} = \rho_A (a/r)^6$ beyond, the chameleon charge in the deep-screening tail regime is $Q_{\mathrm{tr}} = \tfrac{1}{2}(4\varepsilon_{\mathrm{tr}})^{3/4}$, against $Q_{\mathrm{sph}} \simeq 3\varepsilon_{\mathrm{sph}}$ for the sphere, and the symmetron near-zone charge is $Q_{\mathrm{tr}} \simeq 1.014\, s^{-3/2}$, against $3/\kappa^2$. Both decay more slowly, exactly because the tail is never fully screened and always contributes to the exterior monopole. With the physical pion mass the tail is exponentially cut off and the enhancement becomes logarithmic, but the recomputed exclusion regions still extend further into the strongly coupled regime, for instance the hydrogen 1s\u20132s chameleon bound at the dark-energy scale reaches couplings 1.8 orders of magnitude weaker than for a uniform-sphere proton (2.4 in the chiral limit), overtaking the muonium reference. The paper's conclusion is prescriptive: future analyses of single-nucleon probes should use the Skyrmion profile rather than the hard sphere.
-
The central claim is that a holographic dark energy model with infrared cutoff $1/L^2 = -\alpha R + \beta Q$ yields a consistent cosmological evolution that covers all major epochs. Using the Friedmann equation and the conservation of matter, the authors obtain a first-order differential equation for the dark energy density parameter $\Omega_{\rm DE}$, which admits an exact closed-form solution. From this solution they compute the equation-of-state parameter $w_{\rm DE}$ and the deceleration parameter $q$, and show that for suitable parameter choices the universe transitions from deceleration to acceleration at $z\approx 0.45$, that $w_{\rm DE}$ can cross the phantom divide, and that the model can asymptotically approach de Sitter behavior. They also consider an interacting scenario with dark matter and include radiation to demonstrate the full thermal history.
-
The paper's central claim is that the evidence for dynamical dark energy is not one number: it depends on the parameterization chosen for $w(z)$, on which datasets are included, and on whether the model comparison is Bayesian (Bayes factor) or frequentist (AIC, DIC). With DESI BAO DR2 and DES-Dovekie supernovae (with or without gamma-ray bursts, and always with redshift-space distortions), the thawing quintessence form $w(z) = -1 + \delta(1+z)^{-\gamma}$ is the only model preferred over $\Lambda$CDM by all three metrics: moderately by the Bayes factor and positively by AIC and DIC. The CPL form improves the fit, and its equation of state crosses into the phantom regime $w(z) < -1$ at $z \sim 0.5$ at roughly $2$\textendash$3\sigma$, yet the Bayes factor weakly prefers $\Lambda$CDM over CPL while AIC and DIC prefer CPL. The paper resolves this disagreement with an Occam penalty: splitting the evidence as $\ln B \approx \Delta\chi^2_{\min}/2 + (\text{Occam penalty})$ gives penalties of about $4$\textendash$5$ for CPL, $3$\textendash$4$ for AT and only $1$\textendash$2$ for TQD, so identical fit gains are outweighed or not depending on the metric. GEDE, with its dark-energy density suppressed at high redshift, is disfavoured by all three metrics, but relaxing the $\beta$ prior to allow negative values flips the AIC/DIC verdict while the Bayes factor stays inconclusive. The authors conclude that a preference from AIC or DIC alone warrants caution, and that whether phantom crossing is required remains open.
-
The load-bearing claim is that the parameter $E_P$, extracted from the relativistic part of the cross-correlation between a bright and a faint galaxy population, is a genuine null test of the weak equivalence principle and is sensitive to the dark-matter coupling in a controlled way. In this model $E_P$ has a closed form: with a linear potential, constant couplings, and $Y(k)\to1$, it becomes $E_P=1-2\beta_{\mathrm{dm}}x-2\beta_{\mathrm{dm}}(\beta+\beta_{\mathrm{dm}})\,[1-(1-4\beta^2)x^2-y^2+4\beta x]^{1-\gamma}$ in the doubly coupled case, reducing to the simpler coupled-dark-energy expression when $\beta=0$. Since $E_P\to1$ whenever $\beta_{\mathrm{dm}}\to0$ even if $\beta\neq0$, any deviation from unity signals equivalence-principle violation. Projecting this expression, together with the coupling-dependent expansion rate, onto redshift-binned SKA2 constraints gives $\sigma_{\beta_{\mathrm{dm}}}=0.044$ (22%) and $\sigma_{\Omega_{m0}}=0.011$ (3.5%) in the coupled-dark-energy model, with DCDE errors of 430%, 130%, and 24% when both couplings are free.
-
On the paper's own terms, the discovery is that a Hubble-dependent non-conservation scale turns cold dark matter non-conservation into a complete engine for the late-time dark-energy phenomenology. Because the trace-free field equations leave the combination $R + 8\pi G T$ free, a non-conserved cold dark matter fluid forces the reconstruction $\Lambda_{\rm eff} = \frac{1}{4}(R + 8\pi G T)$ to evolve according to $\dot\Lambda_{\rm eff} = -8\pi G\,\Gamma(H)\rho_c$; interpreting $\rho_{\rm DE} = \Lambda_{\rm eff}/8\pi G$ with $p_{\rm DE} = -\rho_{\rm DE}$ yields an effective equation of state $w^{\rm eff}_{\rm DE} = -1 + \Gamma(H)\rho_c/(3H\rho_{\rm DE})$, so the sign of $\Gamma(H)$ decides quintessence-like versus phantom-like behavior and a sign change produces a smooth phantom-divide crossing at finite redshift. Model I, with $\Gamma = \xi_0 H$, has the exact solution $\rho_c \propto a^{-3+\xi_0}$ and reproduces the quadratic running-vacuum law $\rho_{\rm DE}(H) = \rho_{\rm DE,0} + \frac{\xi_0}{8\pi G}(H^2 - H_0^2)$ in the dust-only limit, while Model II, with $\Gamma = H_0[\xi_0 + \xi_1(1 - H_0/H)]$, makes non-conservation negligible relative to expansion at early times and permits reversals in the direction of energy exchange. The unified dynamical-system analysis exhibits a radiation past attractor, saddle matter-era configurations, and a continuous family of de Sitter critical points that are transversely attracting when $\Gamma(H_\star)/H_\star < 3$. Confronted with Pantheon+, cosmic chronometers, and DESI DR2, both models give $\chi^2$ values below $\Lambda$CDM, an AIC slightly favoring the thermogravity scenarios, a BIC still favoring $\Lambda$CDM, and a mild preference for positive present-day non-conservation ($\xi_0 = 0.131^{+0.068}_{-0.069}$ in Model I). The paper is also explicit that the size of the geometric contribution is fixed by an integration condition ($\lambda_0 = 1 - \Omega_{m0} - \Omega_{r0}$) rather than predicted, so thermogravity reframes the cosmological-constant problem rather than solving it.
-
On the paper's own terms, the central claim is that including a scalar-Gauss-Bonnet coupling generalizes the rapid-turn regime of two-field inflation and dark energy in three linked ways. First, the slow-roll relation becomes $$\varepsilon_V \approx \frac{2\$varepsilon^{2}$}{2\varepsilon+\delta_1} + \frac{1}{2}\left(\frac{\sqrt{2\varepsilon+\delta_1}\,\eta_\perp}{3}+f_N $H^{2}$\right)^2,$$ which reduces to the familiar $\varepsilon_V\approx\varepsilon(1+\eta_\perp^2/9)$ when $f$ is constant, but otherwise lets $\varepsilon_V$ be of order one even for non-large turning rates. Second, the Hessian components of the potential acquire Gauss-Bonnet corrections, $V_{;TT}\approx 3H^4 f_{;TT}+H^2\eta_\perp^2$ and $V_{;TN}\approx 3H^4 f_{;TN}+3H^2\eta_\perp$, and the component $f_{;TT}$ that earlier work treated as negligible is in fact non-negligible even in a slow-turn slow-roll regime. Third, a generalized consistency condition relates the field-space metric, the potential, and the coupling. These results imply the Gauss-Bonnet term can drive or enhance rapid turning on its own, can affect the amount of turning needed to resolve the $\eta$-problem, and, for dark energy, permits a phantom equation of state under the sign conditions $\delta_1>0$, $\delta_2<1+2\varepsilon$ or $\delta_1<0$, $\delta_2>1+2\varepsilon$, with dark matter crucial even in the absence of direct interaction.
-
On the paper's own terms, the discovery is that the chameleon branch (NMCQ+) of nonminimally coupled quintessence is preferred over $\Lambda$CDM by the current data, and that its mechanism for doing so is the modified dark matter dilution rather than a phantom field. For the baseline combination the model gives a best-fit improvement $\Delta\chi^2_{\rm min}=-7.25$, a $68\%$ coupling $\beta=0.186^{+0.047}_{-0.038}$, a $68\%$ upper limit $n<0.0703$ on the inverse-power-law index, and a joint $3.3\sigma$ preference for a nonzero interaction when both transfer directions are sampled together. The best-fit trajectory produces an effective dark energy equation of state that crosses $w=-1$ from below to above at $z_{\rm cross}\simeq0.487$, close to the crossing near $z\simeq0.5$ inferred from DESI DR2 BAO with CMB and supernova data, while the intrinsic field equation of state satisfies $w_\varphi\ge-1$ throughout. The improvement over $\Lambda$CDM is comparable for the opposite coupling branch (NMCQ$-$, $\Delta\chi^2_{\rm min}=-7.84$), and the two directions receive nearly equal posterior weight, so the data's preference is for a nonzero dark-sector interaction more strongly than for a specific direction.
-
On its own terms, the paper establishes that an effective scalar–photon-fluid coupling, evolved dynamically rather than imposed as a constant dilution anomaly, predictably reshapes recombination and the linear spectra: the photon temperature departs from the standard $T_0(1+z)$ law, the last-scattering surface moves to higher redshift, the acoustic scale shifts by roughly 3.4% for the illustrative fiducial model, and the matter power spectrum acquires a scale-dependent response with a sign change near $k \simeq 0.091\,h\,\mathrm{Mpc}^{-1}$. When those predictions are passed through the Planck 2018 temperature, polarization, and lensing likelihoods at a fixed six-parameter baseline with $A_{\mathrm{Planck}}$ profiled, the best sampled interacting cosmology, $(\epsilon_i, \sigma) = (4\times10^{-5}, 0.10)$, is slightly disfavored relative to the separately evaluated interaction-off $\Lambda$CDM reference, $\Delta\chi^2 \approx +0.30$. The authors' stated conclusion is that the model produces identifiable cosmological signatures but no observational preference for the interaction, and that the comparison is conditional rather than a general model-selection result.
-
In entropic cosmology, cosmic acceleration is attributed to horizon thermodynamics rather than to a dark-energy component, and the paper's contribution is to test three previously proposed driving terms in one unified effective-fluid framework. Assuming matter and radiation are separately conserved, the two entropic functions $f(t)$ and $g(t)$ are tied by $df/dt + 2Hf = 2Hg$, so each choice of driving term $g(t) = \beta_1 H^2$, $\beta_1 H^2 + \beta_2 \dot H$, or $\beta_1 H^2 + \beta_2 \dot H + \beta_3 H^4$ determines a definite expansion history $E^2(z)$. Fitting those histories to Pantheon+, DESI DR2 BAO, and Planck 2018 CMB distance priors pushes $\beta_1$ to about 0.7--0.9 in all three models and leaves $\beta_2$ consistent with zero within $1\sigma$, while the data place $\tilde\beta_3$ at order $10^{-9}$. The theoretical filters tell a different story: early-universe consistency requires the entropic term to vanish at high redshift, $|\beta_1| \ll 1$ for Model A and $|\beta_1 - 2\beta_2| \ll 1$ for Model B, and the TCC surrogate $q_\infty \ge 0$ requires $\beta_1 \le 0$ for Model A and $\beta_2 \ge \beta_1$ for Models B and C. Consequently the data-favored and theory-favored parameter regions do not fully overlap: Model A, the simplest, cannot satisfy both filters at once, Model B shows the mildest tension while remaining statistically competitive with $\Delta\mathrm{AIC} \approx 2$, and Model C's extra term does not improve the fit.
-
The paper's positive claim, stated in its conclusions, is that the extra-radiation model with lepton asymmetry can still alleviate the Hubble tension without adopting a helium abundance below the standard BBN prediction: $N_{\rm eff}=3.27\pm0.16$, $\xi=0.017\pm0.011$, $H_0=69.24\pm0.92\,{\rm km\,s^{-1}\,Mpc^{-1}}$, with the Gaussian tension against the SH0ES value $73.17\pm0.86$ falling from $5.74\sigma$ in $\Lambda$CDM to $3.12\sigma$. The mechanism is a two-lever compensation: extra radiation speeds weak freeze-out and raises helium, while a positive $\xi$ acts through the charged-current weak rates that set the neutron-to-proton ratio, lowering helium; the radiation energy added directly by $\xi$ is negligible ($\Delta N^{\xi}_{\rm eff}\simeq4\times10^{-4}$), so the asymmetry works almost purely through BBN weak rates. The complementary result is negative for EDE: the BBN likelihoods tighten the 68% upper limit on the EDE fraction from $0.017$ (without BBN) to $0.012$, because the higher baryon densities favored by EDE lower predicted deuterium, and the small $\xi$ allowed by helium does not compensate deuterium. The paper then shows that a positive light-quark-mass variation during BBN, $\delta_q=0.0072\pm0.0044$, increases deuterium and slightly lowers helium, relaxing the EDE bound to $f_{\rm EDE}<0.019$ and raising $H_0$ to $68.85$ (tension $4.02\sigma$), while leaving the extra-radiation constraints nearly unchanged; it also finds that allowing lepton asymmetry relaxes the 95% upper limit on the summed neutrino mass from $0.0640$ to $0.0777$ eV in the extra-radiation model. The authors state that the quark-mass constraints are conditional on the adopted nuclear response coefficients, which are fixed rather than marginalized.
-
The paper proposes a dedicated dark-universe observatory whose single instrument delivers, in every exposure, four-band optical imaging ($g,r,i,z$, $0.15''$ PSF) and contiguous $R\approx100$ NIR ($1.1$–$1.8\,\mu$m) spectra via a linear variable filter, with the two channels split by a dichroic and no moving spectral elements. Because wavelength is encoded in focal-plane position, one survey scan assembles the full griz set and the roughly 48-channel spectrum of every source, so spectral information begins at the moment of detection with no follow-up loop. On this basis the paper forecasts $\sigma(w_p)\approx0.03$ with FoM $\approx80$–$180$ for dark energy from weak lensing, angular BAO/AP, and NIR-standardized Type Ia supernovae; $\sigma(S_8)\approx0.008$ for structure growth; a self-interacting-dark-matter bound $\sigma/m\lesssim0.02$ cm$^2$ g$^{-1}$ from more than a thousand merging clusters; primordial-black-hole fractions $f_{\mathrm{PBH}}\lesssim10^{-2}$ from microlensing; and a dark-siren $H_0$ approaching the 1% level once next-generation gravitational-wave detectors are paired with KST's host catalog. The load-bearing structural claim is that these forecasts are set by the systematics chain, and that KST's chain is internally measured: four-band shear with a thermally stable diffraction-limited PSF under $m<10^{-3}$, LVF mean redshifts to $\sigma(\Delta z)=0.002(1+z)$, and a NIR supernova floor near 0.010 mag. The paper also argues that the 1.4 m aperture is itself an optimum for this architecture: survey grasp scales as $E\propto D^{-1}$ while the density of cleanly resolved galaxies saturates once the PSF reaches the faint-galaxy scale, so the science-yield rate peaks near $D\approx1.4$ m.
-
Working on the principal Lambert-W branch with $0.5<b<1$, selected through phase-portrait analysis because it approximately reproduces the standard matter-dominated trajectory before reaching a de Sitter attractor, the paper constrains the model with CMB-SPA (Planck, ACT, SPT), DESI DR2 BAO, and PantheonPlus data. The headline results are $b = 0.573^{+0.016}_{-0.060}$ and $H_0 = 70.67^{+0.31}_{-0.35}\,\mathrm{km\,s^{-1}\,Mpc^{-1}}$ (68% CL) for the full combination with the SH0ES calibration, with the torsion equation of state crossing the phantom divide at $z\simeq1.29$. The paper establishes an asymmetry: on CMB data alone the model fits slightly better than $\Lambda$CDM ($\Delta\chi^2 = -2.24$) and reduces the SH0ES tension from $5.13\sigma$ to $2.34\sigma$, yet once BAO and supernova distances are included the combined fit is worse ($\Delta\chi^2 = +7.27$) and the Bayesian evidence weakly favors the standard model. It also notes the model never reduces to $\Lambda$CDM for any parameter value, only to GR without a cosmological constant in the $\beta=0$ limit. The conclusion, stated plainly, is that the extra parameter buys a higher $H_0$ but not the low-redshift flexibility needed to match the full dataset.
-
Stated on its own terms, the paper's claim is that Hubble expansion acts as a persistent repulsive force on wide galactic orbits, and that omitting it biases merger predictions. In the high-tangential-velocity ΛCDM runs, the merger is delayed relative to the Newtonian runs by roughly 1 to 7.5 Gyr, and for the lowest masses with tangential velocities at or above ~66.5 km s$^{-1}$ the pair separates instead of merging, a qualitatively new outcome in which the final fate itself depends on whether expansion is included. The effect traces to a single term: in physical coordinates the simulation reduces to Newtonian gravity plus a repulsive central acceleration $\Omega_\Lambda H_0^2 r$ (or its $w_0w_a$ analog), whose amplitude grows linearly with separation. At a post-pericenter apocenter of ~800 kpc this repulsion is ~18% of the gravitational attraction, near the de Sitter scale $r_\Lambda \sim 1.4$ Mpc where dark energy dominates, so low-mass, high-velocity cases sit at a watershed. The $w_0w_a$CDM runs show a weaker delay because, under the adopted parametrization, the dark-energy equation of state crosses zero at $a \sim 1.87$ and the acceleration shuts off. The authors are explicit that the strength of the headline result coincides with the breakdown of their model: the isolated-pair idealization is no longer clearly valid beyond ~1.2 Mpc, and the $w_0w_a$ parametrization is extrapolated tens of Gyr beyond its $z<2.5$ calibration.
-
On the paper's own terms, the discovery is that a scalar–one-form–two-form system with constant $BF$ coupling is not merely consistent but highly structured. Retaining the lapse in the homogeneous reduction keeps the Friedmann constraint alive, and the local duality between each one-form/two-form pair and a Proca field fixes the degree count at twelve: the scalar, the metric, and three pairs each carrying three physical modes. For every background satisfying $I>0$, $J>0$, $m\ne 0$, $H_c\ne 0$, and $U = V + \rho_H > 0$, the exact finite-wavenumber quadratic actions have positive-definite kinetic matrices, constraint inversions with no finite-$k$ pole, and luminal principal characteristics. Within the exponential family $I = I_0 e^{\alpha\phi/M_{\rm Pl}}$, $J = J_0 e^{\beta\phi/M_{\rm Pl}}$, $V = V_0 e^{-\lambda\phi/M_{\rm Pl}}$ with $\alpha+\beta = \lambda$, there is an exact fixed point with $a \propto t^2$, density fractions $\Omega_\phi = 89/108$, $\Omega_F = 17/108$, $\Omega_H = 1/54$, stable within the homogeneous system, with bounded normalized linear perturbations at every fixed finite comoving wavenumber. Choosing the effective potential $W = a^3 V = W_0(c_1 Q_1 + c_2 Q_2)^2$ — stated explicitly as an ansatz, not a consequence of Noether's theorem — the zero-boundary determining equations yield canonical potentials that are hyperbolic-squared or exponential with couplings $I \propto V^{1/3}$, $J \propto V^{-1/3}$; the causal type of the potential coordinate separates the branches, and Branch III integrates exactly with Bessel functions, reaching $a \propto t^{4/3}$, $w_{\rm eff}\to -1/2$, with the total form fraction decaying as $T^{-2}$.
-
The central claim is that the constant-coupling mass-varying neutrino model, where neutrinos couple to a quintessence field with exponential potential and coupling, can be robustly simulated in a fully relativistic regime. Contrary to earlier expectations that this model catastrophically forms neutrino lumps, the authors find that lumps are highly dynamical and spinning, and can dissipate due to the cosmon’s oscillations. The effective dark energy equation of state oscillates about w = -1, allowing phantom crossing without modifying gravity. The code also computes backreaction from large-scale nonlinearities, leading to expansion histories compatible with ΛCDM while producing deviations in the lensing potential and anisotropic stress at levels potentially observable with future data.
-
The central claim is that, using a Gaussian-process reconstruction tailored to the friction function α_M(z) (rather than directly parameterizing the distance ratio Ξ(z)), the GWTC–5.0 catalog provides no evidence for deviations from general-relativistic GW propagation. The reconstructed α_M(z) and the derived Ξ(z) are both consistent with zero and unity, respectively, at all redshifts supported by the data. Moreover, after calibrating the GP prior to match the integrated prior width of a standard Ξ_0–n model, the GP posterior contracts by a factor of 5.2 compared to 2.8 for the parametric model, and the GP posterior covariance reveals a higher effective rank (r_eff=1.56 vs 1.05), indicating that the non-parametric model allocates posterior uncertainty across more independent redshift-dependent directions than the approximately rank-one Ξ_0–n model.
-
The central claim is that axionic quintessence in perturbative heterotic compactifications is viable in principle, but only through aligned instanton charges rather than a single large cycle. The paper derives the 4d axion effective theory from 10d heterotic string theory, including kinetic terms, Green-Schwarz couplings with topological coefficients $n_i$, one-loop gauge kinetic functions $f^{1\text{-loop}}=(S\pm n_i T^i)/(4\pi)$, and the worldsheet-instanton potential $V=-\sum_A \Lambda_A^4\cos(q_A\cdot b+\delta_A)$ with $\Lambda_A^4\propto e^{-2\pi q_A\cdot v}$, then extracts the geometric requirements for an ultralight eigenstate. An exploratory scan over favorable toric Calabi-Yau hypersurfaces with $h^{1,1}=2,3,4$ finds zero controlled configurations satisfying the single-large-cycle criterion, the volume bound $V\lesssim 25$, and the $\alpha'$-control cuts among 4,305,000 sampled directions; the closest candidate, the resolved degree-12 hypersurface $X_{12}\subset \mathbb{P}^4_{1,1,2,2,6}$, gives $\tau_{\min}\simeq 0.9$ at $V\simeq 25$, just outside the control region. In the minimal scheme the benchmark also demands $n_1=77$ to reach $f_{\text{eff}}\simeq 0.4\,M_p$, violating the hidden-sector positivity bound. The aligned configuration $q_H=(3,1,0)$, $q_L=(4,1,0)$ at cycle volumes $(10.8,1.5,1.8)$ with $s=21.5$ and $n_i=1/4$ resolves this: it yields $m_4\simeq 2.7\times 10^{-60}M_p$, $f_4\simeq 0.35\,M_p$, $\alpha_v^{-1}=25.0$, and $\alpha_h^{-1}\simeq 18.0$. The paper presents this as an effective-theory benchmark under explicitly stated stabilization assumptions, and stresses that the required suppression of competing instantons, all 55 audited alignment candidates have lower-action classes with nonvanishing projection on the light direction and nonvanishing genus-zero Gopakumar-Vafa invariants, must come from bundle-dependent Pfaffian zero modes, which it does not construct.
-
The central claim is that shift-symmetric Horndeski gravity with $L=\tfrac{1}{2}R+K(X)-G_3(X)\square\phi$ can cross the phantom divide in a healthy theory. The construction reduces the cosmology to an autonomous two-dimensional system in the variables $x_k$, the kinetic contribution to the dark-energy density, and $x_g$, the braiding contribution (the share coming from the scalar-tensor mixing term); shift symmetry then supplies a conserved charge that fixes the braiding exponent, $m=(2n-1)/4$, where $n=XK_X/K$. The divide $w=-1$ is the parabola $Q=x_g^2+2nx_g+8n^2x_k=0$, and the paper proves that the flow derivative $Q'$ evaluated on the curve is $Q'|_{Q=0}=-\frac{3(2n+1)}{8n}s(s-2n)^2$, which vanishes only at the two fixed points, the early matter point $M$ and the de Sitter attractor $\mathrm{dS}$. At every other point of the curve the flow has a component across it, so a trajectory that reaches $w=-1$ passes through rather than being reflected, and each branch acts as a one-way door. Two mechanisms then realize the crossing: a monomial theory with $n=3/4$ launched near the braiding point, which crosses once and settles into de Sitter, and the evolving sigmoid $n(X)$ of Eq. (5.1) with $0<n(X)<1/2$, which yields one or two crossings while preserving early matter domination. The paper further diagnoses the old no-go: it followed from setting the conserved charge to zero, the late-time de Sitter limit, which restricts motion to the straight line $x_g=-2nx_k$ that only touches the parabola at de Sitter.
-
The central claim is that the distance scale of the Pantheon+ supernovae can be transplanted to gamma-ray bursts through a Bayesian neural network that reconstructs the distance modulus $\mu(z)$ and whose predictive uncertainties are empirically recalibrated, not merely asserted. The Amati relation $\log_{10}(E_{\rm iso}/{\rm erg}) = b + a\log_{10}(E_{\rm p,i}/300\,{\rm keV})$ is fitted only at $z\le 1.5$, where the reconstruction is best constrained, and then used to assign distances to 264 bursts, producing a Hubble diagram extending to $z\simeq 9.4$ that inherits the Pantheon+ distance scale and its covariance structure. The two standard calibration likelihoods give different slopes — D'Agostini $a=1.67$, Reichart $a=2.13$, with the Reichart prescription defining the baseline diagram — and the intrinsic scatter dominates individual distance errors. The paper's robustness diagnostics, including shared-covariance propagation, leaving out the three highest-redshift supernovae, and a redshift-evolution term $\gamma$ consistent with zero, move posterior medians by at most $0.040\sigma$; a truncation-corrected Efron-Petrosian test finds marginal isotropic-energy evolution that can occur along the relation rather than orthogonal to it. Combined with Pantheon+, 34 Hubble-parameter measurements, and DESI BAO, the GRB diagram tilts the model comparison toward dark-energy extensions — the Akaike criterion prefers the Chevallier-Polarski-Linder parametrization while the Bayesian criterion prefers constant-$w$ dark energy — which the paper reads as improved phenomenological fits, not evidence for evolution of the equation of state. Idealized future catalogues of 1500 standardized GRBs would, under the adopted uncertainty model, discriminate $\Lambda$CDM from CPL in all 50 simulated realizations.
-
The authors test the cosmological background of Schouten–Codazzi gravity, which introduces one extra parameter $\Omega_{\gamma_0}$ beyond $\Lambda$CDM. Using three different Type Ia supernova samples (Union3, DES-SN5YR, Pantheon$+$) together with DESI DR2 BAO and a compressed CMB likelihood, they find that the three-probe combinations all give $\Omega_{\gamma_0}=0$ within $1\sigma$: $0.029^{+0.048}_{-0.045}$, $-0.007^{+0.033}_{-0.034}$, and $0.033\pm0.039$ for the three supernova compilations respectively. The Bayes factors favor $\Lambda$CDM with $\Delta\ln\mathcal{Z}$ between $-2.54$ and $-1.13$. The two-probe combinations without CMB or without BAO show a positive shift, but this disappears when all probes are combined. The authors conclude that at the background level, current data do not require the SCG modification.
-
The central claim is that the biverse volume-normalized action $A=(V_F/V_g)S+S_F$ removes an arbitrary cosmological constant from our universe and turns it into a small correction to the curvature of the other universe. Gravity in our universe is quantized not with $\hbar$ but with the rescaled constant $\hbar_q=\hbar V_g/V_F$, so in the limit $V_g/V_F\to 0$ all gravity loops vanish: our metric behaves as a classical field coupled to conventionally quantized matter, and no loop-generated counterterms are produced. The otherwise mysterious fiducial metric of ghost-free massive gravity appears naturally as the pullback of the bulk metric onto our universe, dark energy arises from the graviton mass, and the theory predicts that the tensor-to-scalar ratio is exactly zero, $r=0$, irrespective of the inflationary model.
-
The central discovery is that the dark dimension (DD) scenario, which postulates a micron-scale extra dimension whose size is related to the dark energy (DE) scale, can be realized in an EFT framework only under restrictive conditions. In a minimal 5D model where the radion couples to the 4D Ricci scalar (Brans–Dicke form), the competition between a 5D quintessence potential and Casimir energy fixes the DD size and yields a 4D vacuum energy that scales with the quintessence field as expected from the (A)dS distance conjecture, but this vacuum energy cannot act as DE in a Friedmann–Lemaître–Robertson–Walker universe because the radion equation of motion forces the 4D effective Newton constant to vary inconsistently. The obstacle is removed in a 5D model where the 4D Einstein–Hilbert action is induced by quantum loops from a large number of species (N ∼ O(100)), making the 4D Planck mass effectively independent of the radion. In that model the DD size is determined by the 5D quintessence potential, the 4D DE emerges from the same mechanism, and the resulting cosmology is viable. Alternative 6D models with fixed total internal volume either leave the DD size as an input or require a fine-tuned subtraction to bring the 4D DE down to the observed value, which violates the asymptotic dS swampland conjecture's scaling relations. The authors conclude that a universal low-energy EFT link between the DD size and DE remains elusive.
-
The ΛGfE theory demonstrates that by adding a term proportional to the Lagrange multiplier Λ to the GfE Lagrangian—a term that vanishes on shell—and demanding Weyl stationarity of the total action modulo the volume constraint, one obtains modified gravity equations whose trace-free part is insensitive to vacuum energy density contributions. This is achieved because the Λ term adjusts to exactly cancel the trace of the stress-energy tensor, leaving equations that depend only on the traceless part. In the low-energy, small-curvature limit, these equations reduce to the trace-free Einstein equations, which are known to be unaffected by vacuum energy that would otherwise contribute a large cosmological constant. Beyond that limit, the equations retain two dark energy contributions: the constant Λ and the dynamical Λ_G, the latter emerging from the G-field and vanishing in the weak-field regime.
-
Fitting the joint $w_0w_a$CDM$\xi$ model to Planck TT, DESI BAO, and Pantheon+ gives $w_0=-0.796\pm0.079$, $w_a=-0.56^{+0.34}_{-0.30}$, $M_\nu=0.41\pm0.13$ eV, and $\xi_{\nu 3}=1.14^{+0.18}_{-0.13}$, compared with $M_\nu=0.06$ eV and $\xi_{\nu 3}=0$ in the no-neutrino case. The semi-analytical analysis fixes the CMB acoustic scale and angle-averaged BAO angles and Taylor-expands the response in $w_0$, $w_a$, $M_\nu$, and $\xi_{\nu 3}$, producing Eq. 3, a quadratic relation between $w_a$ and $\xi_{\nu 3}$ whose curves track the MCMC contours. Two parameter sets on the same perturbation curve, one with $w_a=0$ and $\xi_{\nu 3}=1.15$ and the other with $w_a=-0.83$ and $\xi_{\nu 3}=0$, generate CMB temperature spectra with matching peak and trough positions. The paper concludes that the DDE and $\Lambda$CDM$\xi$ models are degenerate at the level of CMB and BAO acoustic scales, so separating them requires constraints beyond the expansion history.
-
The central claim is that the late-time attractor of the axio-dilaton system is a non-geodesic fixed point, not a single-field scaling or geodesic solution. For a dilaton $\phi_1$ rolling in $V(\phi_1) = V_0 [A + (\phi_1/M_P - B)^n] e^{-\lambda \phi_1/M_P}$ and coupled to a massless axion $\phi_2$ through the kinetic function $W = e^{-\xi \phi_1/M_P}$, the attractor has equation of state $\omega_\phi = (\lambda - 2\xi)/(\lambda + 2\xi)$, which approaches $-1$ for any positive slope $\lambda$ provided $\xi \gg \lambda$. The mechanism is that $W_{,\phi_1} < 0$ makes the axion kinetic energy push the dilaton up its potential, against the force $-V_{,\phi_1}$, and this competition stabilises the trajectory; the curvature of the field space, not the flatness of the potential, is what drives acceleration. The non-geodesic points exist and are stable for $\sqrt{6+\xi^2}-\xi \leq \lambda \leq 2\xi$, no physical fixed point is phantom, and for the Albrecht--Skordis ($n=2$) correction a stable de Sitter minimum appears when $A\lambda^2 < 1$, while the non-geodesic point becomes the attractor when $A\lambda^2 > 1$. Against DESI DR2 BAO, binned Pantheon supernovae and Planck 2018 distance priors, the model prefers steep slopes and strong couplings, $\lambda \gtrsim 7$--$9$ and $\log_{10}\xi \gtrsim 2.5$ at 95% confidence, with a degeneracy between the parameters and a small improvement over $\Lambda$CDM.
-
The paper establishes that three phenomenological interacting dark matter-dark energy models (interacting vacuum, phantom dark energy with $w_0 < -1$, and quintessence dark energy with $w_0 > -1$) all produce a statistically significant preference for negative effective neutrino masses when the positivity prior is removed. None of the resulting 95% confidence intervals accommodates the minimum neutrino mass required by terrestrial oscillation experiments ($\sum m_\nu \geq 0.057$ eV for normal ordering). Allowing the effective neutrino mass to vary increases the significance of a non-zero interaction to approximately $3\sigma$ in the phantom case, but the preferred masses remain negative. The improvement in $\chi^2_{\min}$ and weak Bayesian support for some models does not translate into a viable neutrino sector, because the central values of $\sum m_{\nu,\mathrm{eff}}$ are negative and the 95% upper limits stay below the oscillation floor.
-
The paper's central claim is that the $r_d$-agnostic strategy — leave the sound horizon free, let BAO constrain $r_d H_0$ and $\Omega_m$, and let CMB and galaxy lensing constrain $\Omega_m h^2$ — reaches useful precision with the next generation of surveys. For fixed $\Sigma m_\nu$, the forecast is $\sigma(H_0)=1.3$ km/s/Mpc with the baseline priors, $0.69$ km/s/Mpc with an informative Gaussian prior on $A_s$, and $1.5$ km/s/Mpc with the least informative prior set; allowing $\Sigma m_\nu$ to vary loosens $H_0$ to roughly $1.3$–$1.8$ km/s/Mpc while bounding the neutrino sum at 95% CL near $0.36$ eV. The paper also predicts a 25–30% reduction in $\sigma(H_0)$ relative to the current-data analysis it extends, with the biggest gain occurring when $\Sigma m_\nu$ is free. The boundary of the claim is set by its validation tests: mock data from a modified-recombination model are recovered without bias under all three priors, while mock data from an early-dark-energy model are not, with the pipeline returning $H_0\simeq 69.8\pm 1.0$ km/s/Mpc against an injected value of $71.45$ km/s/Mpc.
-
The central claim is that the combined Planck, DESI DR2 BAO, and DESY5 supernova data do not exhibit the kind of model misspecification this test can see: the observed posterior-disagreement statistic lies near the median of the CPL reference distribution, with $p_{\rm global}=0.5243$. The supporting methodological discovery is that a calibrated, alternative-free model check can be built from the disagreement between two amortized neural posterior estimators trained on the same simulations, with a two-ensemble calibration converting the minimum of three distance-based p-values into a valid global p-value. Controlled experiments show that the test detects departures through their differential effect on the two estimators: it is genuinely powerful for a rapid-transition equation of state (100% rejection at the strongest injection) but nearly blind to the smooth JBP and BA parameterizations at $(w_0,w_a)=(-0.8,-1.0)$, which are flagged only 1.5% of the time, a form-dependence the authors acknowledge as room to improve sensitivity.
-
The paper shows that a phenomenological parametrization unifying early- and late-time expansion history, while providing a better statistical fit to current cosmological data than ΛCDM, cannot significantly raise the Hubble constant when the early-time background is restricted to an analytic radiation–matter transition. The central discovery is that even with additional freedom in the late-time expansion (via parameters η1, η2, η3), the sound horizon at last scattering remains nearly unchanged, and the Hubble constant inferred from CMB+BAO+SNe data (with or without the SH0ES prior) stays far below the local measurement. This implies that the Hubble tension cannot be solved by merely adjusting the matter-like density or late-time dynamics; an additional localized feature in the pre-recombination expansion history, such as that provided by early dark energy, is required.
-
The paper establishes that among four phenomenological interacting dark energy models with a constant equation-of-state parameter w, the model where the interaction term Q = β H_0 ρ_de (IwCDM3) is the most strongly supported by current cosmological data. For the CMB+DESI+DES-Dovekie dataset, the coupling parameter β is measured to be -2.$52^{{+0.80}}$_{-0.54}, deviating from zero at 3.2σ, and w = -1.$510^{{+0.170}}$_{-0.120}, deviating from -1 at 3.0σ. The model yields a Bayesian evidence ln B = 4.3 relative to ΛCDM, indicating moderate support. Crucially, the effective dark energy equation of state derived from the interaction evolves with redshift, crossing the phantom divide w_eff = -1, and closely matches the behavior of the w0waCDM model that DESI originally favored. This shows that the DESI preference for evolving dark energy can be interpreted as a signature of dark-sector interactions rather than an intrinsic time dependence of dark energy.
-
The authors study the quadratic model f(Q,C) = $η1Q^{2}$ + $η2C^{2}$ + η3CQ in a flat FLRW universe. They reformulate the equations of motion into an autonomous dynamical system for two families of affine connections (Class II and Class III). In both cases, they identify a fixed point corresponding to accelerated expansion with weff = -1+ε, which is stable for all ε < 2/3, making it a universal attractor. This shows that accelerated expansion is an intrinsic geometric effect of the f(Q,C) sector, requiring no dark energy component. Additionally, for Class II, a dust-like matter-dominated fixed point exists and satisfies energy-momentum conservation; for Class III, the matter solution involves a mixture of pressureless dust and stiff fluid, indicating a matter-geometry coupling. The authors introduce an ε-parametrization to quantify deviations from -1 and coin the terms 'Quintaverse' (ε>0) and 'Phantiverse' (ε<0) for the two regimes.
-
The central discovery is that a cyclic $\mathbb{Z}_N$ symmetry that protects an ultralight scalar from radiative corrections does not preclude appreciable mass variation of dark matter, provided the relic populations of the $N$ symmetry-related partners are unequal. Such inequality arises naturally from the scalar-induced mass splitting during nonrelativistic freeze-out, without explicit $\mathbb{Z}_N$ breaking, because the scalar background spontaneously breaks the symmetry. When number-conserving conversions maintain internal equilibrium after the total abundance is fixed, the relative abundances are given by equilibrium weights (Equation 15) that depend only on the fermion temperature and the scalar background at conversion decoupling. The resulting effective mass modulation (Equation 21, $m_{\text{eff}}^\psi(\theta) \simeq m_{\text{sym}}^\psi(\theta) - \epsilon M [I_1(q)/I_0(q)] \cos(\theta-\theta_i)$) yields $O(\epsilon)$ sensitivity when $q \gtrsim 1$, i.e., when conversions decouple deep in the nonrelativistic regime.
-
The paper's central claim is that the non-Gaussian component of tomographic lensing convergence maps carries recoverable information about the dark-energy equation-of-state parameters $w_0$ and $w_a$ that auto- and cross-power spectra alone do not capture. In the authors' forecast, a data vector combining power spectra with Minkowski functionals, moments, peak counts, minima counts, and wavelet scattering coefficients reduces marginalized $1\sigma$ uncertainties by 24.3% on $w_0$, 28.0% on $w_a$, 41.9% on $\Omega_m$, and 45.8% on $\sigma_8$ relative to the power-spectrum baseline, using all source planes $z_s = \{0.5,1,1.5,2,2.5,1100\}$, where $z_s = 1100$ denotes CMB lensing. The gains are not uniform in origin: galaxy-lensing tomography is what first breaks the $w_0$–$w_a$ degeneracy, and only then do the non-Gaussian statistics add their contribution, whereas CMB lensing contributes mainly to $\Omega_m$ and $\sigma_8$ and responds only weakly to dark-energy evolution. The authors also identify the information carriers: Minkowski functionals and moments give comparable power and outperform peak, minima, and scattering statistics, with the derivative moments $S_1 = \langle \kappa^2 \nabla^2 \kappa \rangle$ and $K_1 = \langle \kappa^3 \nabla^2 \kappa \rangle$ doing most of the moment work, and the joint vector beats every individual statistic.
-
This is an end-to-end injection-recovery forecast. The authors take the DESI DR2 best-fit 29-bin w(z) reconstruction—with w₁≈+0.83 at the lowest redshifts, H₀=57.365, Ωm=0.423—as the fiducial cosmology, generate mock binary-neutron-star catalogs under ET-D sensitivity with instrumental error σ_inst≈2dL/ρ and weak-lensing scatter σ_lens=0.05zdL, keep events above a network SNR of 8, and then rebuild the expansion history from the recovered distance–redshift relation. Reconstructed via a Horndeski-motivated regularized 29-bin fit and decomposed by PCA, the GW3000 catalog alone yields a covariance-weighted reconstruction SNR of 7.170 and lnB₇,Λ=20.874 against ΛCDM; the CPL fits give Wilks significances of 3.309σ and 6.317σ for 1000 and 3000 events using GW data alone, and 7.710σ and 11.033σ when combined with DESI DR2 BAO, BBN, and the compressed Planck PR4 acoustic-scale constraint. The effective dimensionality stays small—N_eff≈3.8–5.2 for GW-only catalogs and about 5.4–6.3 in the combined analysis—so the information lives in a handful of broad, correlated modes, with w>−1 at low redshift, w<−1 around z≈0.7–1.0, and an effective CPL projection in the Quintom-B sector.
-
The core discovery is that a linear differential equation of third order arises from the holographic density ansatz $\rho_H = 3M_p^2 [\alpha H^2 + \beta \dot{H} + \gamma \ddot{H}/H + \eta (1/H) d/dt(\ddot{H}/H)]$ when recast in the variable $x=\ln a$ with $E=H^2$. The conditions $A_m=A_r=1$ fix two parameters, leaving the solution $H^2 = H_0^2 (\Omega_{m0}a^{-3}+\Omega_{r0}a^{-4}+\sum C_i a^{\lambda_i})$, where the characteristic exponents $\lambda_i$ satisfy $\eta\lambda^3+\gamma\lambda^2+\beta\lambda+2(1-\alpha)=0$. Two roots can fall between $\lambda=-3$ (matter) and $\lambda=-4$ (radiation), enhancing expansion around recombination, while a slightly positive root ($0.15\lesssim\lambda_1\lesssim0.30$) acts as a phantom component that raises the late-time $H_0$ when the acoustic scale is fixed, albeit strongly limited by BAO distances.
-
The paper demonstrates that the axion dark energy potential V(ϕ) = V0[1 + p cos(ϕ/f1)] with 0 < p < 1 yields a fixed-point structure where the radiation- and matter-dominated solutions are saddle-type critical lines, ensuring they serve as transient epochs, while the de Sitter solution at the potential minimum (Ωϕ = 1, wϕ = −1) is a stable future attractor. The analysis further reveals a transition from a stable node to a stable focus depending on the dimensionless combination F = κf1 and p, explaining how the dark-energy equation of state can exhibit damped oscillations without crossing the phantom divide.
-
The paper establishes a systematic relation between the minimum cosmic age allowed in the analysis and the inferred departure from ΛCDM in the CPL parametrization. As the lower bound on t₀ is raised, the allowed region in the (w₀, wₐ) plane shifts toward (w₀, wₐ) = (−1, 0), and the covariance-weighted Mahalanobis distance decreases. This behavior is not driven by a single threshold but is a continuous trend: expansion histories that lie far from ΛCDM tend to produce younger universes, so requiring older cosmic histories preferentially retains models closer to the standard cosmological constant.
-
The paper claims that the DDELenS simulation suite is the first to provide full N-body simulations and ray-traced weak lensing convergence maps across 128 cosmologies in the $w_0 w_a$CDM parameter space, covering $\Omega_m \in [0.28,0.35]$, $\sigma_8 \in [0.74,0.90]$, $w_0 \in [-1.0,-0.5]$, and $w_a \in [-1.5,0.0]$, with 10,000 realizations per cosmology (and 100,000 at the fiducial) at six source redshifts for tomographic and cross-correlation analyses. The pipeline uses the N-body gauge to consistently incorporate relativistic effects from massive neutrinos, photons, and dark energy, and the simulations are validated against theoretical predictions: matter power spectrum and lensing convergence power spectrum agree with HMcode-2020 to within ~5% on quasi-linear scales, and the halo mass function agrees with Tinker et al. (2008) within ~10%. The suite is designed for simulation-based inference of dynamical dark energy parameters using non-Gaussian statistics such as peak counts, Minkowski functionals, and field-level neural posterior estimation.
-
The central discovery is that the CMB birefringence angle of (0.264 +/- 0.058) degrees can be accounted for by a massless pseudo-scalar field with equation of state w=1, without requiring the field to be dark matter or dark energy. The field's energy density redshifts as (1+z)^6, implying a kination era in the early universe. The coupling constant must lie in a narrow window: g_phi_gamma between 8.9e-14 and 5.0e-13 $GeV^{-1}$, bounded from below by BBN and from above by quasar observations.
-
The authors propose that dark matter consists of dark 't Hooft–Polyakov monopoles and dark energy is an ultralight axion coupled to a dark gauge field. The axion induces an electric charge on the monopoles via the Witten effect, giving a dyon mass $m_D(n,\theta) = m_M + \gamma m_M (\alpha_D^2/2)(n-\theta/(2\pi))^2 + \cdots$ and thus a direct DM–DE interaction. The dominant one-loop correction from a single dyon would produce a Coleman–Weinberg potential far larger than the observed dark energy scale. However, because the dyon states form a monodromy tower indexed by integer $n$, the full sum over all states—performed via Poisson resummation or worldline instanton techniques—yields an exponentially suppressed potential with action $S = 2\pi m_\chi / m_{\chi\varphi}$. In the minimal scenario this fixes the DM–DE coupling strength at $\alpha_D^2/2 \approx 0.1\%$. The paper argues that if a second axion (with its own decay constant and rotation history) populates higher-charge dyons asymmetrically, the effective coupling $n_e \sim 1/\alpha_D \sim 30$ can raise the interaction to the percent level, producing an apparent phantom equation of state that yields a raw $\chi^2_{\rm BAO}$ as low as $\sim 9$ for DESI DR2 data while explaining the $\beta \approx 0.30^\circ$ cosmic birefringence signal with anomaly coefficients $|c_\gamma| \sim \text{few}$.
-
On the paper's own terms the central discovery is this: the Schouten-Codazzi field equations, with a Codazzi tensor engineered to the spacetime's Segre type (the classification of a symmetric two-tensor by eigenvalue degeneracies and the causal character of its eigenvectors), admit exact generalized Vaidya and LTB solutions whose departure from general relativity is fully captured by two free functions. For the Vaidya metric $ds^2 = -f\,du^2 + 2\,du\,dr + r^2 d\Omega^2$, the field equations force $f = 1 - 2M(u)/r + \gamma(u)\,r + \lambda(u)\,r^2$, the radiation density keeps the GR value $\kappa\nu = 2M_{,u}/r^2$, and the Misner-Sharp quasilocal mass becomes $M_{MS} = M(u) - \tfrac{1}{2}\gamma(u)\,r^2 - \tfrac{1}{2}\lambda(u)\,r^3$, with a nonvanishing $\gamma$ producing a scalar-curvature singularity at $r=0$. For the LTB dust metric the shell evolution is the cubic $Y_t^2 = 2M/Y - K - \gamma Y - \lambda Y^2$ with $\lambda = -\Lambda/3$, equivalently $Y_t^2 = Q(Y)$ for $Q(Y) = 2M - K Y - \gamma Y^2 + (\Lambda/3) Y^3$, while the dust density retains the standard relation $\kappa\rho = 2M_{,r}/(Y_{,r}Y^2)$. Because $\gamma<0$ adds a positive contribution to $Y_{,tt}$ and $\gamma>0$ shortens the collapse time, the Codazzi sector can mimic dynamical dark energy; the case $\gamma=0$, $\lambda = -\Lambda/3$ recovers $\Lambda$LTB, so the cosmological constant itself is reinterpreted as an integration function of the geometric sector.
-
The central claim is that future TianQin standard siren measurements will leave the current preference for dynamical dark energy intact regardless of which background cosmology produced the data. After fitting $\Lambda$CDM and the four-bin wbin model to the DESI DR2+CMB+DESY5 combination, the authors feed the mean parameters into the same massive-black-hole-binary simulation to make TQSS1 ($\Lambda$CDM-based) and TQSS2 (wbin-based). The catalogs' luminosity distances differ by at most $1.48\%$, with the low-redshift part tracing the input Hubble constant and the $1.02\%$ gap at $z\simeq0.85$ attributed to the genuine equation-of-state difference; this smallness is presented as evidence that the catalogs are comparable probes. Yet the two catalogs push in opposite directions when combined with the real data: TQSS1 lowers the dynamical-dark-energy significance from $3.0\sigma$ to $2.3\sigma$, while TQSS2 raises it to $3.8\sigma$. The paper further reports that switching from light (pop III) to heavy (Q3nod) black hole seeds changes only the number of simulated events and the tightness of contours, not the conclusions, and it acknowledges that the mock catalogs cannot independently assess the Hubble tension because $H_0$ is an input.
-
The paper's central claim is that, for a string network in the scaling regime, each cosmological era contributes a loop number density obeying $t^4 n_p(l,t) \propto (l/t + \Gamma G\mu)^{3\beta-4}$, with $\beta = Ht = \tfrac12, \tfrac23, \tfrac13$ in the radiation, matter, and kination eras. Because the exponent is largest for kination and the kination scaling solution has the smallest correlation-length parameter ($\xi_k = 0.155$ versus $\xi_r = 0.271$), meaning a denser network of long strings, a kination era produces a larger loop number density than the standard radiation era, while matter and vacuum epochs dilute loop production. Combining this with the accumulated emission over cosmic history gives the high-frequency hierarchy $\Omega_{\mathrm{KIN}}^{\mathrm{GW}} > \Omega_{\Lambda\mathrm{CDM}}^{\mathrm{GW}} \gtrsim \Omega_{\mathrm{eDOF}}^{\mathrm{GW}} > \Omega_{\mathrm{EMD}}^{\mathrm{GW}} > \Omega_{\mathrm{IVD}}^{\mathrm{GW}}$. The frequency where each scenario departs from $\Lambda$CDM is set by the scale at which the new epoch ends, $f_{\mathrm{NP}} \sim (10\,\mathrm{Hz})\times \Lambda_{\mathrm{NP}}/\mathrm{TeV}$, because the relevant gravitational waves were emitted at redshift $\sim 10^4\, z_{f\mathrm{NP}}$ by loops created at the end of the new-physics epoch. The authors conclude that TeV-scale kination, EMD, and IVD eras could be told apart from $\Lambda$CDM and from each other by next-generation ground-based detectors, that MeV-scale epochs leave signatures for space-based interferometers, and that the nHz band is blind to all of them.
-
On the paper's own terms, the discovery is that the Bayesian model-selection preference for dynamical dark energy is an artefact of the uniform prior, not a robust conclusion demanded by the data. While the marginalised posteriors on $w_0$ and $w_a$ remain robust to the prior choice across BAO + CMB and all four supernova samples, the Bayesian evidence is highly sensitive: every dataset combination that favoured $w_0 w_a$CDM under uniform priors is either reversed or pushed strongly toward $\Lambda$CDM when the Jeffreys prior is applied. The Kullback-Leibler divergence for $w_0 w_a$CDM rises, for BAO + CMB + DESY5, from $D_{KL} = 30.38$ to $40.04$, demonstrating that the Jeffreys prior's Occam penalty is what swings the evidence. A robustness check shows the Jeffreys-prior evidence is essentially insensitive to the truncation bounds (changing by $|\Delta \ln Z| \le 0.1$), whereas the uniform-prior evidence shifts by up to $|\Delta \ln Z| = 1.36$; so the $\Lambda$CDM preference is not an artefact of the bounds imposed to make the prior proper. The paper is explicit that this does not establish the Jeffreys prior as uniquely favoured, nor does it rule out evolving dark energy; it establishes that the evidence claim should not be interpreted independently of prior specification.
-
The paper's central claim is that in the Higgsed phase of MacDowell–Mansouri pre-geometric gravity the vacuum energy is quantized, $\Lambda_k \sim M_{Pl}^2/k$, with the integer $k$ labeling vacua separated by barriers in a periodic gravi-Higgs potential. Because the Gibbons–Hawking entropy of each sector is $S_{dS} \sim k$, the number of available microstates grows as $e^k$, and the total tunneling rate $\Gamma_{k\to k'} \sim e^{k' - S_E}$ is dominated by this entropic factor whenever the Euclidean bounce action grows no faster than $k$. The ratio of upward to downward transition rates then becomes $r_k \to e^2$, making the ordered sequence $\Lambda_1 > \Lambda_2 > \cdots$ the most probable path, with the cosmological constant decreasing monotonically toward the asymptotic attractor $\Lambda_\infty = 0$. The author concludes that the observed vacuum energy is not fine-tuned but is the finite-age remnant of relaxation from an initial Planckian vacuum, with $k \sim 10^{120}$ today, and that the irreversibility of the relaxation supplies a thermodynamic arrow of time.
-
The paper establishes that the U-shaped warps of 3968 edge-on disk galaxies in the Sloan Digital Sky Survey, when modelled as a response to a Yukawa fifth force acting only on dark matter, show no statistically significant correlation with the predicted fifth-force field. This absence bounds the fifth-force strength relative to Newtonian gravity to ΔG/G_N < 7.7×$10^{{-4}}$ at 95% credibility for λ_C ≥ 8 Mpc, and < 1.7×$10^{{-3}}$ after inflating for systematic uncertainties. The limit tightens with increasing force range, and the authors argue it applies conservatively to the massless (infinite-range) limit. The bound is independent of the scalar field's cosmological evolution and is complementary to cosmological constraints.
-
The central claim is that the three inequivalent symmetric-teleparallel connection branches allowed by cosmological symmetry in scalar-nonmetricity gravity are not just mathematical possibilities but produce distinct, observationally differentiable background cosmologies. The data favor Connection I over ΛCDM when the DES Y5 supernova sample is combined with CMB and DESI DR2 BAO data, while Connections II and III yield expansion histories similar to ΛCDM but with more parameters, and are not favored. The fact that the same action and potential lead to different posterior structures and model-comparison results for the three connections demonstrates that the choice of affine connection has observable consequences at the background level.
-
The central claim is that the black hole information paradox can be avoided if the Universe has a finite lifetime (a Big Crunch) and the Page time of every existing black hole exceeds that lifetime. The paper shows that the smallest astrophysical black holes have Page times far exceeding any plausible finite Universe lifetime ($10^{58}$ Gyr for a solar-mass black hole), but hypothetical light primordial black holes could have shorter Page times. By demanding that even the lightest primordial black hole never reaches its Page time, the author derives a lower bound on black hole mass of about 7.5e14 g (for a 35-Gyr Universe). Consequently, no black holes with mass below ~$10^{15}$ g can exist, and the Universe is configured so that the paradox is never realized.
-
Coupled thawing quintessence, with either a cosine (pseudo-Nambu-Goldstone) or a double-exponential hilltop potential, can reproduce the qualitative features of the DESI DR2 evolving dark energy signal: an effective equation of state today w0 > -1 and a past crossing of the phantom divide without a phantom field. The coupling amplitude is constrained to |β|<0.073 at 95% credibility, and its posterior is bimodal with modes at β≈±0.05. Only the negative mode (β<0) produces the crossing, at z_cross≈0.9–1.0, while the positive mode does not. The data favour the negative branch by Δχ²≈1, meaning the crossing is viable but not required. The two potentials give statistically identical constraints on β but differ in the equation of state they can reach: the cosine is consistent with the present-day slope of the CPL fit, while the double-exponential cannot produce steep enough evolution.
-
The central claim is that the nonlocal parameter $\gamma$ in the path-B branch of the Maggiore–Mancarella model is constrained by the combined DESI DR2 BAO, Planck 2018 CMB, and Pantheon+ supernova data to be $\gamma < 0.00495$ at 95% confidence, making the model effectively indistinguishable from $\Lambda$CDM. The paper derives a local consistency relation $w_a = -\alpha_{\text{NL}} (1 + w_0) + O((1 + w_0)^2)$ with $\alpha_{\text{NL}} \approx 0.686$, which shows that departures from $\Lambda$CDM drive $w_0$ away from $-1$ and $w_a$ more negative in a fixed direction that does not match the degeneracy direction of the CPL constraints. Consequently, even when standard cosmological parameters are varied, the one-dimensional trajectory predicted by the model does not cover the observationally preferred region, forcing the posterior to accumulate at the $\Lambda$CDM endpoint.
-
The unified BDE-BDM model derives both dark energy and the entire dark matter from a single confining supersymmetric SU(3) gauge theory with six massless flavors. Gauge-coupling unification, assuming the dark and Standard Model couplings meet at ~$10^{16}$ GeV, predicts the condensation scale Λ_c = $34^{{+16}}$_{-11} eV without any cosmological input. The present dark-energy density, the time-varying equation of state, and the timing of the transition are then fixed by Λ_c through the field equations, leaving no free parameter in the dark-energy sector. The model fits DESI DR2 BAO, Planck CMB, CMB lensing, and supernova data, recovering Λ_c = 43.67±0.13 eV — consistent with the prediction — and yields w0 = -0.9262±0.0005, with a thawing quintessence that stays above the phantom divide. Relative to ΛCDM, it improves the fit by Δχ² ~5-9 and is favored by a Bayes factor ln B ≈ 4.1-4.8, while the two-parameter w0waCDM model shows no Bayesian preference over ΛCDM.
-
Starting from the Einstein equations for a Bianchi type I metric with a perfect fluid $\rho$ interacting with a vacuum component $V$ through a phenomenological transfer $Q_\mu = Q u_\mu$, the paper reduces the dynamics to a closed four-dimensional autonomous system in the expansion rate $\Theta$, the total shear $\sigma$, the density $\rho$, and the vacuum energy $V$, together with the constraint $\Theta^2/3 - \sigma^2 = \rho + V$. For the linear coupling $Q_1$, the fluid and vacuum equations integrate explicitly in terms of the average scale factor $S = (abc)^{1/3}$, and the constraint becomes a mechanical identity $y^2 + U(S) = 1$ whose potential is a combination of powers of $S$ plus a shear term proportional to $S^6$. The central positive claim is that for parameters with $E_1 < 0$ and $\xi > 2$ the potential has an infinite barrier as $S \to 0$, so orbits descend from a contracting de Sitter repeller, bounce at a finite minimal $S$ where $y = 0$ and $y' \neq 0$, and then expand toward a de Sitter attractor; near the bounce both $\rho$ and $\rho + V$ turn negative, the localized energy-condition violation the authors argue is what evades the singularity theorems. The central negative claim concerns the nonlinear coupling $Q_2$: the surfaces $x = 0$ and $v = v_\Lambda$ are invariant, the physical domain $x > 0$, $v \geq 0$ is preserved, and the constraint forces $y^2 \geq 3(x + v) > 0$, so the expansion rate never vanishes; in this anisotropic setting, unlike the isotropic case, $Q_2$ admits no bouncing trajectories for physical matter, and the only physical equilibria are a contracting/expanding de Sitter pair.
-
On its own terms the paper establishes that the DESI DR2 BAO measurements, when combined with CMB distance priors and Pantheon+ supernovae, do not favor a common f(T)-gravity departure from ΛCDM. The evidence is the contrast among three models sharing the same data: for the power-law model the mild BAO+CMB shift ($b = -0.08 \pm 0.07$) disappears after supernovae are included, so the full combination is consistent with ΛCDM; the exponential model retains $b = 0.35^{+0.08}_{-0.07}$ in the full fit, but the improvement in $\chi^2$ only offsets the extra parameter in AIC ($\Delta\mathrm{AIC} = -0.10$) while BIC prefers ΛCDM ($\Delta\mathrm{BIC} = +5.29$); and the hyperbolic-tangent model, which has no ΛCDM limit, performs markedly worse ($\Delta\mathrm{AIC} = 19.17$). The paper's argued conclusion is that parameter-level deviations must be separated from statistical model preference, and that the DESI-inspired departures from ΛCDM are not a generic feature of f(T) gravity.
-
The central claim is Theorem 1: under the conditions $M^2_\varphi = M^2_\chi \equiv M^2$ and $\gamma_\varphi = \gamma_\chi \equiv \gamma$, the coupled, linearized, small-excursion system admits the closed form $X(x) = X_i - \beta(X_i) + \mathrm{Im}(u(x))$ and $\Phi(x) = \Phi_i - \tan(\Phi_i) + \mathrm{Re}(u(x))$, where $u(x) = C e^{-i(k+\gamma/2)x}\,{}_1F_1(A; 3p; 2ikx)$, with $k = \sqrt{M^2 + \gamma^2/4}$, $A = \frac{3p}{2k}(k + \gamma/2)$, and $\beta(X_i) = \frac{\sin X_i}{n + n\cos X_i - 1}$. The proof complexifies the two real equations into $u'' + (3p/x + i\gamma)u' + M^2 u = 0$ and substitutes $u = e^{i\kappa x} f(z)$ with $z = 2ikx$, which maps the equation exactly onto Kummer's equation $z f'' + (3p - z)f' - A f = 0$; regularity at $x = 0$ discards the Tricomi solution, and the initial field values fix the complex constant. In the weak-coupling limit $\gamma \to 0$ the formula reduces to a fractional-order Bessel function, recovering the single-field structure, while for $\gamma \to \infty$ the fields stay pinned near their initial values. Numerical integration of the original coupled equations corroborates the closed form wherever the excursions $\delta X$ and $\delta\Phi$ remain small relative to their initial values, with the $n = 1$ potential agreeing longest and very large coupling giving essentially exact agreement.
-
On the paper's own terms, the central finding is that combining linear-stability analysis with physical-admissibility conditions and observational calibration singles out a single late-time endpoint in this particle-creation cosmology. The candidate accelerating attractors include the formally de Sitter-like dark-matter branches $P_7$ and $P_8$, which have $\omega_{\rm tot}=-1$ and $q=-1$, but at all three best-fit parameter sets they lie outside the physical compact interval $0\le u\le 1$: $u_7<0$ for $P_7$ and $u_8>1$ for $P_8$. The dark-energy-dominated point $P_{10}=(0,0,0,1)$ satisfies acceleration ($\omega_{\rm de}<-1/3$), linear stability, and physical admissibility simultaneously for PPS+BAO+OHD, Union3+BAO+OHD, and DES-DOV+BAO+OHD. Since $P_{10}$ sits at the compact boundary $u=1$ where $H\to 0$, the asymptotic evolution is a quintessence-like power law with a decreasing Hubble rate, not a finite-$H$ de Sitter state. Posterior-level analysis shows 92%--97% of the MCMC samples support this stable accelerating endpoint, and the reconstructed expansion reproduces the radiation--matter--acceleration sequence with a transition redshift $z_{\rm tr}\simeq 0.71$--$0.73$.
-
The paper establishes that in a KS spacetime containing radiation and a generalized Chaplygin gas, the anisotropy parameter β(t) does not oscillate or diverge but instead stabilizes at a finite constant value after a finite time, irrespective of the initial conditions. Because the KS metric in Misner form reduces to the FRW metric when β is constant (up to coordinate redefinition), this means the model isotropizes. The isotropization occurs before recombination (ts ≈ $10^{12}$ s for observationally fitted parameters), consistent with the high isotropy of the cosmic microwave background. The accelerated expansion required to drive isotropization is provided by the GCG’s negative pressure, while radiation dominates at early times. The authors verify the behavior by systematically varying the GCG parameters α, A, C and the radiation constant Cr, showing that β approaches a constant in every expansive solution.
-
The central thesis of this review is that the upcoming multi-band gravitational-wave network—from pulsar timing arrays, through LISA and deci-Hertz concepts, to next-generation ground-based detectors like Einstein Telescope and Cosmic Explorer—will enable transformative science across astrophysics and cosmology. This includes precise measurements of black hole and neutron star mass and spin distributions, distinguishing binary formation channels, using standard sirens to measure the Hubble constant and dark energy, and conducting precision tests of general relativity. The author contends that the dramatic increase in sensitivity and frequency coverage will inevitably lead to unexpected discoveries, given the historical pattern of new observational windows.
-
The paper establishes that the KL divergence, the expected log-likelihood ratio under the alternative, does not by itself determine discovery potential when the alternative is a continuous parameter family. Detectability is governed by the distribution of the log-likelihood ratio, the size of the parameter space searched, and whether the null lies on a boundary. For Gaussian data, the log-likelihood ratio follows either an exactly Gaussian distribution (mean shift) or a generalized chi-square distribution (covariance shift), and its cumulants are derived. An operational detection statistic—peak prominence—is calibrated against null simulations to set false-positive rates. In four examples (toy Gaussian, cosmic topology, tensor-to-scalar ratio, dark energy), the detection probability at a given KL divergence varies with the model family and detection threshold, and parameter estimates from detected peaks are biased upward by selection effects.
-
The paper reports a systematic offset between XMM and eROSITA measurements of intracluster medium properties for a sample of 114 galaxy clusters matched across the DES redMaPPer, XMM Cluster Survey, and eRASS1 catalogues. Using power-law fits with fixed slope, the authors find that XMM temperatures are 28±7% higher than eROSITA temperatures, and soft-band (0.5–2 keV) and bolometric (0.1–10 keV) luminosities are 19±6% and 34±7% higher, respectively. When the slope is allowed to vary, the temperature offset shows no strong dependence on temperature, but the luminosity offsets show a luminosity‐dependence: the offset decreases for more luminous clusters. The X-ray–richness scaling relations (L_X–λ and T_X–λ) have consistent slopes and scatter between the telescopes, but differ in normalisation, confirming that the telescopes can be used jointly after scaling.
-
The paper demonstrates that LLM agents can efficiently navigate a large space of modified-gravity Lagrangians and identify models that fit late-time expansion data competitively with flat ΛCDM. The key finding is that all eight independently initialized searches returned at least one candidate with a lower nominal AIC than ΛCDM, and seven of them recovered the well-studied quadratic-quintessence class. This shows that the combination of grammar-based generation, deterministic physical validation, and LLM-guided selection can automate the discovery of viable theoretical models in cosmology, while also revealing a limitation: the recovered models are almost always those that reduce to general relativity plus a single canonical scalar field.
-
The paper establishes that when a common power-law tilt ε in the distance duality relation is allowed to vary freely in both ΛCDM and w0waCDM, the improvement in fit of w0waCDM over ΛCDM is substantially reduced for every combination of DESI DR2 BAO and four state-of-the-art supernova compilations. Quantitatively, the reduction fraction f_red (the fraction of the Δχ² improvement that disappears when ε is freed) ranges from 57% for Pantheon+, through 43% for DES-Dovekie and Union3, down to 18% for Union3.1. In terms of Gaussian-equivalent significance, the preference for dynamical dark energy drops from 1.8σ to 1.0σ for Pantheon+, from 2.2σ to 1.5σ for DES-Dovekie, from 2.7σ to 1.9σ for Union3, and from 2.0σ to 1.8σ for Union3.1. The tilt ε itself is weakly constrained and consistent with no violation at the ~1-2σ level, but its inclusion absorbs a significant portion of the signal that would otherwise be attributed to w0 and w_a. The result is robust to alternative parametrizations of the magnitude redshift trend and is more effective at reducing the signal than adding spatial curvature alone.
-
The central discovery is a finite-time upper bound on the accumulated vacuum energy suppression D_Λ(N) = log_10(ρ_{Λ,i}/ρ_Λ(N)) for power-law metric f(R) = f_0 R^n with n>2. The bound combines the asymptotic attractor rate (below 4/(3 ln 10) ≈ 0.579 decades per e-fold) with an integral constraint from the scalar field velocity range 0 ≤ x ≤ 4|β|. For the reference initial state (Ω_{Λ,i}=0.99, Ω_{kin,i}=$10^{{-4}}$), the bound gives D_Λ(60) < 47.7 decades, ruling out a 60-decade reduction within 60 e-folds. Additionally, the paper identifies a stability exchange at λ = 4|β| that leaves the relaxation rate continuous, and shows through numerical integration that the f(R) ∝ $R^{{2.14}}$ trajectory yields D_Λ(60) ≈ 7.53 decades, with about 3.27 decades due to transient excess above the asymptotic rate. Brans-Dicke and field-dependent coupling examples further demonstrate the role of the coupling-weighted kinetic history.
-
The claim is stated on the paper's own terms in Section 3.1: the mean intergalactic dispersion measure responds as $\delta\ln\langle{\rm DM}_{\rm IGM}\rangle \simeq \delta\ln H_0 + \delta\ln\Omega_{\rm b}$, so FRB data alone carve out a broad band of roughly constant $\Omega_{\rm b} H_0$ in the $H_0$--$\Omega_{\rm b}$ plane. DESI DR2 BAO, taken without any external calibration of the drag-epoch sound horizon $r_d$, constrain the combination $H_0 r_d$ and produce a band that is nearly transverse to the FRB band. Their intersection is compact: $H_0 = 71.4^{+2.2}_{-2.5}$ km s$^{-1}$ Mpc$^{-1}$, $\Omega_{\rm b} = 0.0503^{+0.0025}_{-0.0021}$, and $r_d = 142.4 \pm 4.8$ Mpc (Equation 20, Table 2). The inferred $H_0$ lies between the Planck value $67.36 \pm 0.54$ and the SH0ES distance-ladder value $73.04 \pm 1.04$ km s$^{-1}$ Mpc$^{-1}$, consistent with both at 1.7$\sigma$ and 0.6$\sigma$, and $\Omega_{\rm b}$ agrees with the joint BBN+CMB determination at better than 1$\sigma$. The paper also reports that the $H_0$--$\Omega_{\rm b}$ result is barely changed when supernova samples are added, whereas the dark-energy parameters $w_0$, $w_a$ remain dominated by BAO and supernovae.
-
The central claim is that the simplest relaxation law for the dark sector — constant fractional relaxation of the logarithmic density ratio, $X' = -nX$ with $X = \ln(r/r_\infty)$ — is exactly equivalent to the dark-energy equation of state $w = w_0 a^{-n}$. Because of the exact identity $X' = -3w$, this one law generates a complete history: dark energy is exponentially suppressed at early times, crosses the phantom divide once from below at $a_c = |w_0|^{1/n}$, reaches its maximum density exactly at that crossing, and then dilutes toward a state in which both components scale as $a^{-3}$ with $w \to 0$. The future is consequently decelerating expansion, no big rip, and no cosmological event horizon. Against DESI DR2 BAO, Planck distance priors, and three supernova compilations, the family prefers nonzero $n$ over a constant equation of state at a nominal $2.6$–$3.6\sigma$, with the benchmark $n = 1/2$ matching the two-parameter CPL fit with one fewer dark-energy parameter. The paper further shows that near equilibrium the same $n$ governs perturbations: long-wavelength composition disturbances relax at the background rate, small-scale modes are damped sound waves with $c^2_{\rm comp} = n/3$, and common matter–dark-energy perturbations cluster like dust — while proving that a pointwise constitutive closure cannot remain gradient-stable across the full negative-pressure past, which delimits where additional pressure-response physics is needed.
-
The authors claim that a systematic 3x3 grid of holographic dark energy models, each combining one of three generalized entropies (Barrow, S3, S5) with one of three infrared cut-offs (Granda–Oliveros, plus two extensions adding gamma $H^{{-2}}$ or gamma \ddot{H}/H), reveals a sharp dichotomy: the \ddot{H}/H term (cut-off 3) is well constrained and substantially improves the fit for Barrow and S5 entropies, whereas the $H^{{-2}}$ term is never constrained by current data. The overall best fit is Barrow entropy with cut-off 3, but the Barrow deformation parameter $\Delta$ is compatible with zero, so the classical Bekenstein–Hawking area law is not falsified. This indicates that the extra dynamical freedom required by the data resides in the infrared cut-off rather than in a modification of the entropy.
-
The authors show that logarithmic (FRSlog) and exponential (FRSexp) Finsler-Randers-Sasaki geometries can both provide viable phenomenological descriptions of the late-time background expansion when constrained by CC, BAO, and Pantheon+ data. Under the baseline uncalibrated fit, both models yield H0 near 68 km/s/Mpc, close to ΛCDM. When the SH0ES local prior is added, H0 rises to about 71.6 km/s/Mpc in both cases. Despite similar background behavior, the two models produce distinct higher-order cosmographic signatures: FRSlog gives a positive present-day jerk (j0 ≈ 0.4–0.6) and remains near the ΛCDM statefinder fixed point, while FRSexp exhibits a strongly negative jerk (j0 ≈ −1.1 to −1.3) and a much larger excursion in the snap and lerk parameters. Both models yield a quintessence-like effective dark-energy equation of state (wDE,0 > −1) throughout the relevant redshift range.
-
The central claim is that a single canonical field can account for both anomalies at once. In this model the neutrino mass is fixed to its present-day value $M_{\nu,0} = 0.06$ eV — the normal-ordering floor — while being allowed to be smaller in the past; the reconstruction $M_{\nu,\rm eff}(z)$ then falls below that floor at earlier epochs, so the sub-floor values reported by cosmological analyses and the oscillation floor measured on Earth refer to different epochs of the same mass history. At the same time, an observer who analyzes the expansion history under the standard assumption of constant neutrino masses will attribute the missing energy to dark energy and read off an apparent equation of state that crosses $w_{\rm app} = -1$ at $z_{\rm cross} = 0.90^{+0.28}_{-0.13}$ in 96% of the posterior histories, while the underlying field obeys $w_\phi \geq -1$ and the null energy condition. The quantitative machinery is a coupled background-plus-perturbations implementation in a Boltzmann solver fitted to CMB, DESI DR2 BAO, and supernova data, which returns a preference for a positive coupling ($\beta = 5.4^{+3.9}_{-5.3}$) with $\beta = 0$ at the edge of the 95% credible region, stable standard cosmological parameters, and the same qualitative histories under both DES Y5-Dovekie and Pantheon+ supernova samples.
-
The paper's central claim is that the marginalised posterior in full-shape analyses is distorted by a 'Laplace term', $\log\det F(\Theta)$, which measures how the volume of the nuisance-parameter space varies across cosmological parameters. In ${\rm Mpc}/h$ units this term is steep and strongly $h$-dependent, tilting the posterior toward high $h$ and low values of the dark-energy equation of state; in ${\rm Mpc}$ units it is nearly insensitive to $h$. Because the likelihood is exactly invariant under the unit change while the priors are not, since dimensional nuisance parameters such as $c_\ell$, $c_{\rm nlo}$, $a_{\rm vir}$ and $N^P_{2\ell}$ scale with different powers of $h$, the difference between the two conventions is driven entirely by prior-volume effects. The authors support this with a direct experiment: fixing $h$ to its fiducial value makes ${\rm Mpc}/h$ and ${\rm Mpc}$ analyses agree almost perfectly, and a ${\rm Mpc}/h$ run whose non-linear scale and number density are fixed in ${\rm Mpc}$ units reproduces the ${\rm Mpc}$ posteriors, showing that what matters is on which parameters the priors are imposed, not the nominal units of the fit. On this basis they conclude that ${\rm Mpc}$-based priors alone restore unbiased constraints across $\Lambda$CDM, $K$CDM, $w$CDM and $w_0w_a$CDM, with the $\sigma_{12}$-based reparametrisation outperforming $\sigma_8$ because $8\,{\rm Mpc}/h$ is itself $h$-dependent while $12\,{\rm Mpc}$ is not.
-
Testing six representative extensions of ΛCDM against the full Planck 2018 likelihood versus a ΛCDM-calibrated Gaussian compression on the vector (1/θ*, ωm, ωb), the paper finds a clean sector-based divide. Dark-energy models that preserve matter scaling reproduce the full analysis to better than about 4% in posterior width along every direction. But every model that changes how dark matter redshifts — interacting dark energy (ξ < 0), decaying dark matter into dark radiation, and ΛwDM — breaks the compression, with center displacements up to S_CC ≈ 2.1, strongly anisotropic deformations (A_aniso up to ~1.9), mode reordering, and genuine rotation of principal directions. The mechanism is the loss of a degeneracy the full likelihood sees: varying the new parameter while adjusting the present-day dark-matter density can keep the recombination-era abundance, and hence the CMB spectra, nearly fixed; the compressed likelihood, calibrated on the ΛCDM relation between ω_m and recombination, penalizes exactly that compensating shift. The paper argues the relevant question is not when new physics acts but whether it preserves the relation between present-day and recombination-era matter abundance.
-
On the paper's own terms, the discovery is a degeneracy: a first-order $\Lambda$LTB geometry described by just one amplitude $m_{0,\mathrm{amp}}$ and one scale $r_0$ produces the same distance-redshift relation that DESI-era data attribute to dynamical dark energy, with best-fit $\chi^2$ differences from flat $w_0 w_a$CDM near zero ($\Delta\chi^2_{w_0w_a\mathrm{CDM}}$ between $-1.49$ and $+1.22$ across the four supernova compilations). All four compilations prefer a shallow overdense core (positive $m_{0,\mathrm{amp}}$), and imposing the one-sided kSZ bound $D^{kSZ}_{\ell=3000} \le 1\,\mu\mathrm{K}^2$ still leaves a viable region with $\sim 10\%$ matter excess and characteristic scale $\sim 300$ Mpc/$h$. The paper stresses that the compensated profile, which turns from overdense to underdense along the past light cone around $z\approx 0.2$, mimics an effective dark-energy equation of state crossing $w=-1$ — the quintom behaviour DESI reports. The conclusion to a fair reader is that weak violation of the cosmological principle can absorb the apparent DDE signal without invoking new dark-energy dynamics.
-
The central claim is that the full $C_\ell^{EB}$ spectrum, not just the overall rotation angle, carries the physical origin of cosmic birefringence. Treating the background four-vector $P^\mu$ in a common Chern-Simons interaction, the authors compute $C_\ell^{EB}$ for a neutrino-asymmetry current $J^\mu=(\Delta n_{\nu_e},0)$ and compare it with late-time axion dark energy and axion dark matter. They find that the neutrino spectrum is suppressed at $\ell\lesssim 20$ like the dark matter case, while dark energy is not, and that the $(1+z)^3$ scaling of the neutrino asymmetry during recombination leaves a different high-$\ell$ shape from axion dark matter. In forecast settings, the scenarios separate with large $\Delta\chi^2$ values at $\ell=51$--$1490$ and $\ell=100$--$2500$; interpreting the current best-fit rotation as a neutrino current yields an effective scale $M\simeq1$ TeV for representative couplings.
-
On the paper's own terms, the central result is that the bispectrum monopole $B_{000}$, appended to the power-spectrum monopole and quadrupole, reduces the marginalised uncertainties on $\omega_{\mathrm{cdm}}$ and $\ln(10^{10}A_s)$ by 9-18% and 8-20% in single-tracer fits, by 15% and 10% for the combined LRG sample, and by 6% and 4% across all tracers, while the bispectrum quadrupole $B_{202}$ adds only a few percent. The improvement is attributed to the bispectrum's ability to constrain the second-order bias parameters $b_2$ and $b_s$, breaking degeneracies between galaxy bias and cosmology that persist in the power spectrum alone. In the $w_0w_a$CDM extension the bispectrum does not meaningfully shrink the allowed dark-energy region, so the main benefit lies in the matter-sector and amplitude parameters rather than in dark-energy evolution.
-
A single field-dependent Gauss-Bonnet coupling, whose magnitude is set dynamically by the vacuum expectation values of three scalars rather than assigned, lets one potential explain both inflationary phases and the late-time dark sector. In the first phase the higher-curvature term dominates the dynamics and drives super-inflation ($\dot{H}>0$), producing a mild blue tilt that accounts for the low-multipole power deficit; the phase ends when the geometrically induced stabilization of the waterfall field fails at a critical field value fixed solely by the combination $(n-1)Q$. The decaying waterfall condensate generates the first radiation, replacing reheating and establishing the thermal bath for the warm phase, where cubic dissipation $\Gamma_\phi\propto T^3$ with shear-viscosity damping makes a quartic potential viable, giving $n_s\approx 0.962$ within the Planck $1\sigma$ band and $r\propto Q_{\rm wi}^{-5/2}$ suppressed to $1.96\times 10^{-3}$–$1.07\times 10^{-2}$. Internal consistency, not the measured spectral index, confines the model to $9.34\times 10^{-3}\le (n-1)Q \le 1.333\times 10^{-2}\,M_p^{-1}$, and the release of a third scalar near matter-radiation equality converts a very small mass scale into a large, galactic-scale Gauss-Bonnet coupling by a logic inverse to the seesaw mechanism. The inflaton then persists as dark energy at late times.
-
On the equatorial plane, axial closed timelike curves are identified with the set where $g_{\phi\phi}<0$, i.e. where the rotational Killing field has negative norm. The paper proves that one additional parameter in each of three modified geometries controls this region. In the tidally charged braneworld black hole the region exists only for positive tidal charge: $P(r)=r^{4}+a^{2}r^{2}+2Ma^{2}r-a^{2}q$ is strictly increasing on $r>0$ with $P(0)=-a^{2}q<0$, so exactly one positive root $r_{\mathrm{CTC}}$ exists for $q>0$ and it lies strictly inside the inner horizon, because $P(r_-)=(r_-^{2}+a^{2})^{2}>0$; for $q\leq 0$ all terms of $P(r)$ are positive and the equatorial region is eliminated. In the Rastall case the horizon itself can never belong to the region since $Q(r_h)=r_h^{2}g_{\phi\phi}(r_h)=(r_h^{2}+a^{2})^{2}>0$, and for dust, dark energy, and perfect dark matter there are finite intervals of the Rastall coupling across which the positive-$r$ region is absent (e.g. $1/3<\kappa\lambda\leq 0.6$ for dust at $M=1$, $a=0.5$, $q=0.8$, $N_s=0.03$). In the charged Johannsen-Psaltis spacetime $g_{\phi\phi}$ depends linearly on the deformation $\epsilon_3$, with the sign of $\partial g_{\phi\phi}/\partial\epsilon_3$ fixed by $r^{2}+2Mr-q^{2}$, and demanding $g_{\phi\phi}>0$ for all $r>0$ yields a window $\epsilon_3^{-}<\epsilon_3<\epsilon_3^{+}$, numerically $-0.0439<\epsilon_3<-0.0124$ for $M=1$, $a=0.5$, $q=0.8$, in which the equatorial region disappears.
-
Using parameter sets drawn from MCMC chains fit to a common dataset (Pantheon+ SN Ia, OHD, DESI DR2 BAO, and Planck 2018 CMB distance priors), the authors evolve the linear matter growth equation under the smooth-dark-energy approximation and compute $f\sigma_8(z)$ and $S_8(z)$ for the HDE, THDE, and BHDE models. They find that the $f\sigma_8$ and $S_8$ curves for HDE and BHDE are nearly indistinguishable, while the THDE curves lie lowest; that THDE yields the smallest $\chi^2$ for both observables ($\Delta\chi^2 = -1.439$ for $f\sigma_8$ and $0.199$ for $S_8$ relative to $\Lambda$CDM, versus $\Delta\chi^2 > 7$ for HDE and BHDE on $S_8$); and that in the $(f\sigma_8, S_8)$ plane only the THDE and $\Lambda$CDM trajectories reach the current 68% CL data region, while HDE and BHDE trajectories overlap and miss it. The central conclusion is that structure formation can distinguish THDE from HDE and BHDE, even though the models are indistinguishable in the background observables $\Omega_{de}$ and $\omega_{de}$ alone.
-
The central claim is that the Raychaudhuri equation, a purely geometric identity independent of any specific gravitational field equations, can be promoted from a constraint to a reconstruction engine for f(Q) gravity. For a constant deceleration parameter, the method yields f(Q)=C1√Q + D Q^(3m/2); the Friedmann constraint fixes the constants and the derivative-viability filter f'(Q)>0, f''(Q)>0 eliminates the homogeneous √Q term, leaving a pure power law f(Q)=D Q^(3m/2) in the viable m>1 domain. For a constant jerk j=1, the $\sinh$ branch gives f(Q)=E+F(√Q+Q). In both cases the reconstructed Lagrangian satisfies the independent Friedmann constraint, the reconstructed forms pass the background derivative-viability test for the constant-deceleration case, and the paper argues this demonstrates that geodesic-congruence dynamics can be used as the central equation of an inverse reconstruction of f(Q) gravity.
-
The central claim is that the combined data prefer a negative glitch parameter $\Omega_g \equiv 1 - G_\mathrm{N}/G_\mathrm{cosmo} = -0.0081 \pm 0.0025$, corresponding to $G_\mathrm{cosmo}/G_\mathrm{N} = 0.9920 \pm 0.0025$, a 0.8% weakening of gravity on cosmological scales. This is a $3.3\,\sigma$ preference, already present at $2.1\,\sigma$ in the primary CMB alone and sharpened, not introduced, by the BAO data. The glitch resolves the $2.1\,\sigma$ CMB–DESI tension in the $(r_d h, \Omega_m)$ plane to $0.2\,\sigma$ by opening up one parameter while keeping the CMB constraint tight, whereas $w_0w_a$CDM reduces the tension only to $1.3\,\sigma$ by making both posteriors individually unconstrained. In the Hořava-like gravity language the mismatch maps to $\lambda - 1 = 0.0054 \pm 0.0017$, a positive kinetic deformation at the half-percent level; the authors stress that this is evidence for the mismatch realized by such theories, not a detection of Hořava gravity as a complete theory.
-
The central claim is that the exact relation $D_2(r) = 3 + \tfrac{d}{d\ln r}\ln\left[1 + \int_0^\infty \Delta^2(k) W_{\mathrm{TH}}(kr)\,d\ln k\right]$ can be compressed, to better than one percent accuracy, into the sharp-cut-off form $D_{2,\mathrm{cut}}(r;\alpha) = 3 - \Delta^2(\alpha/r) / \left[1 + \int_0^{\alpha/r}\Delta^2(k)\,d\ln k\right]$ with a single globally calibrated constant $\alpha \simeq 2.392$. The physical reading is that $D_2$'s deviation from homogeneity is governed by the dimensionless power crossing a moving Fourier boundary, $k_{\mathrm{cut}} = \alpha/r$, normalized by the cumulative power accumulated below that boundary. Nonlinear evolution lowers $D_2$ at small scales and low redshift because it shifts power across that boundary faster than the accumulated integral grows, and the authors attribute the effect to the relative change of the two contributions rather than to the power enhancement alone. The paper states the accuracy claim as better than one percent on $D_2$ itself for the representative linear and nonlinear cases examined, and explicitly notes that the same relative accuracy does not transfer to the smaller departure $3 - D_2$.
-
The central claim is that metastable (radioactively decaying) dark energy on the minimal phantom brane is preferred over ΛCDM given Planck/ACT/SPT CMB data, DESI DR2 BAO measurements, and the DES-Dovekie Type Ia supernova sample. For the three decay prescriptions — exponential decay of the dark energy density (M1), decay into non-baryonic dark matter (M2), and decay into dark radiation (M3) — the authors report positive decay rates $\Gamma/H_0 = 0.362^{+0.094}_{-0.076}$, $0.54 \pm 0.13$, and $0.48 \pm 0.11$ respectively, with improvements over ΛCDM of $\Delta\chi^2_{\rm MAP} = -17.14$, $-18.76$, $-18.22$ and $\Delta{\rm DIC} = -8.66$, $-9.63$, $-6.56$. The load-bearing physics is that the two effects act in opposite directions: braneworld screening drives the effective equation of state phantom at $z \gtrsim 0.4$, while decay drives it non-phantom toward the present, so the combined sector crosses $w = -1$ at $z \simeq 0.3$–$0.4$ even though the metastable component itself remains quintessence-like throughout. The authors stress that this provides a physical mechanism, consistent with DESI observations, for an effective crossing of the phantom divide without introducing a ghostlike phantom scalar.
-
On its own terms, the central claim is that there exists a region of the model's parameter space — small exponential parameter $\lambda \sim 10^{-3}$ to $0.31$ and a small Gauss-Bonnet coupling variable $u \sim 10^{-7}$ to $10^{-2}$ — in which the effective equation of state crosses the phantom divide several times close to the matter-dark energy transition ($N \sim 0$), while the tensor sound speed stays in $0 \le c_T^2 \le 1$, the scalar sound speed stays in $0.5 \le c_s^2 < 1$, the no-ghost quantity $Q_T = 1 - 8H\dot{f}$ remains positive, and the system asymptotically approaches a de Sitter point with $w_{\rm eff} = -1$. The field oscillates around the potential minimum only for a finite epoch, then exits toward the attractor, so the model avoids the future singularity that can plague phenomenological oscillatory parametrizations.