Pith. sign in

REVIEW 2 major objections 5 minor 39 references

Catching the Cosmic Sign Flip: Background and Growth Tests of Smooth Sign Switching Lambda_s CDM

T0 review · 2 major / 5 minor · reviewed 2026-07-10 · grok-4.5

Pith's one-line read Late-time distance and growth data do not prefer a smooth AdS-to-dS sign switch over flat Lambda-CDM, and cannot pin down how sharp the switch is.

desk verdict Clean late-time null on smooth tanh ΛsCDM: DES-SN5YR (Dovekie) + DESI 2024 BAO + compact RSD do not prefer it over ΛCDM, and Δ is prior-dominated; useful baseline, not a tension fix. read the letter →

arxiv 2607.06735 v1 pith:HWGNU7MX submitted 2026-07-07 astro-ph.CO

classification astro-ph.CO
keywords sign-switchingdarkenergyΛsCDMsmoothtanhtransitionDESIBAODES-SN5YRRSDfσ₈modelcomparisonnext-generationforecasts
topics Dark Energy
verification ladder T0 review T1 audit T2 compute T3 formal

The pith

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

The reading

This paper asks whether current late-time cosmology can see a smooth cosmic sign flip: vacuum energy that is negative (AdS-like) at high redshift and positive (dS-like) today. The author replaces the usual discontinuous step with a hyperbolic-tangent transition controlled by a switch redshift z† and a smoothness parameter Δ, then fits DES supernovae (Dovekie recalibration), DESI 2024 BAO distances, and a compact set of RSD fσ₈ points. The joint fits give nearly the same minimum χ² as flat Lambda-CDM; information criteria and Bayesian evidence penalise the extra parameters; and Δ stays prior-dominated, so the data cannot tell a sharp jump from a gradual crossover. The transition sits deep in the matter era (z† around 3.5), where dark energy is only a few percent of the budget and current probes have little leverage. The result is a clean late-time null: background and compressed growth alone cannot resolve the switch that full CMB-plus-BAO analyses invoke for the effective neutrino-mass boundary problem. Forecasts with next-generation supernova, BAO, and CMB-lensing data are offered as the path that could break the degeneracy.

What carries the argument

The normalised hyperbolic-tangent vacuum profile Ω_Λ(z) = (1-Ω_m0-Ω_r0) tanh[Δ(z†-z)] / tanh[Δ z†], which recovers the abrupt signum transition as Δ o ∞ and supplies a continuous effective pressure while still allowing a density zero-crossing; this profile is fitted with free late-time calibration K = c/(H_0 r_d) and a vectorised sub-horizon growth ODE for fσ₈.

What would settle it

A full CMB-plus-BAO-plus-supernova likelihood that drives the posterior on Δ away from the prior (or a next-generation SO + DESI-style BAO + LSST Y3 analysis that recovers a tight peak near the forecast σ(Δ) ≈ 1) would show that late-time data can resolve the transition sharpness after all.

Watch

Extended reading notes

Core claim

Joint DES-SN5YR (Dovekie) + DESI 2024 BAO + compact RSD fσ₈ data do not prefer smooth tanh ΛsCDM over flat ΛCDM: χ²_min values are essentially identical, model-selection statistics penalise the extra parameters, and the smoothness parameter remains prior-dominated (Δ ≈ 24.75 with uncertainties spanning most of the prior), so present late-time data cannot distinguish a discontinuous phase transition from a smooth dynamical crossover.

Load-bearing premise

The claim that a simple sub-horizon growth equation with matter-era initial conditions and seven independent RSD points is a fair proxy for growth even when the effective equation of state diverges at the density zero-crossing.

Share X Bluesky LinkedIn Reddit HN

Signed reviews

No signed human review yet.

Editorial analysis

A structured set of objections, weighed in public.

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

Referee Report

2 major / 5 minor

Summary. The paper generalises the sign-switching cosmological constant model (Λ_s CDM) from a discontinuous step to a smooth hyperbolic-tangent vacuum profile Ω_Λ(z) ∝ tanh[Δ(z† − z)], normalised so that the present-day flatness constraint is exact (Eq. 2). Using DES-SN5YR with Dovekie recalibration, DESI 2024 BAO (with free calibration K ≡ c/(H0 rd)), and a compact seven-point RSD fσ8 compilation, it places background-only and joint background+growth MCMC constraints on (Ω_m0, K, z†, Δ, σ8). The central empirical result is a null detection: χ²_min values are essentially identical to flat ΛCDM (background 1656.15 vs 1656.07; joint 1664.18 vs 1663.95), AIC/BIC and nested-sampling evidence penalise the extra parameters, and Δ remains prior-dominated (Δ = 24.75^{+17.42}_{-17.48}). The authors conclude that current late-time data alone cannot resolve transition sharpness or redshift, and they present Fisher forecasts for next-generation surveys.

Significance. If the null result holds, it cleanly delimits what late-time geometric and compressed growth data can say about AdS-to-dS vacuum transitions motivated by the DESI neutrino-mass boundary tension. Strengths include the free-K setup that isolates late-time geometry without early-Universe priors, analytic marginalisation over SN absolute magnitude, nested-sampling evidence, explicit prior-volume tests on Δ, and an analytic argument that at z† ∼ 3.5 dark energy is only ∼2.5% of the budget so distance and sub-horizon growth observables are insensitive to Δ. The work is a useful baseline rather than a decisive test of the model’s original motivation, which requires full CMB likelihoods.

major comments (2)
  1. §3 and Eq. (8): the joint growth analysis treats seven RSD fσ8 points as independent and solves only a vectorised sub-horizon linear growth ODE with fixed matter-dominated initial conditions at x_ini = −4. The paper itself notes that w_eff diverges at the density zero-crossing (Eq. 5) and that a full Einstein–Boltzmann treatment is needed. While the null claim on Δ is already supported by the background-only analysis and the analytic matter-dominance argument, the joint statement that “current growth data do not break the transition sharpness degeneracy” rests on this phenomenological proxy. The manuscript should either (i) quantify the approximation’s domain of validity near z† (e.g., by comparing to a fluid prescription with c_s² = 1 across a range of Δ) or (ii) more clearly demote the RSD results to a consistency check rather than a joint constraint that informs the headline claim.
  2. §5 and the discussion of neutrino-mass boundary tension: the paper repeatedly motivates Λ_s CDM by its ability, in full CMB+BAO+SN fits, to restore a physical ∑m_ν > 0, yet the present analysis deliberately omits CMB likelihoods and free K. The additional run with a Gaussian K prior (z† = 3.32^{+1.16}_{-1.10}) still leaves the transition deep in the matter era. The central claim is therefore correctly scoped as a late-time null, but the abstract and conclusions should state more sharply that this work does not test the model’s primary claimed resolution of the neutrino-mass anomaly; that requires a full CMB analysis that is outside the present scope.
minor comments (5)
  1. Table 2: the abrupt Λ_s CDM column is blank for the joint (SNe+BAO+RSD) block; a brief note explaining why the abrupt model was not re-run with growth data would avoid reader confusion.
  2. Figure 4 caption and §4: the shaded “main BAO leverage (z ≤ 1.5)” region understates that DESI includes Lyman-α points to z ∼ 4.2; clarifying that high-z BAO errors are large would strengthen the analytic insensitivity argument.
  3. §3: the RSD data vector is listed with effective redshifts and values, but the covariance treatment (“independent effective constraints”) should be stated once in the likelihood equation for reproducibility.
  4. Appendix A: the scalar-field reconstruction is illustrative and useful, but the integral for ϕ(z) (Eq. A3) assumes a real square root; a short remark on the domain where the kinetic term remains non-negative would help.
  5. Typographical consistency: the abstract and body alternate between Λ_s CDM / Λ𝑠CDM and “sign switching” / “sign-switching”; pick one style.

Circularity Check

1 steps flagged · score 1.0 of 10

No significant circularity: central null is an empirical chi^2/AIC/BIC/evidence comparison on public late-time data; tanh is a stated phenomenological ansatz, not a self-derived prediction.

  1. self citation load bearing [Sec. 6 (Forecasts), mock construction paragraph]
    "The supernova mock is generated following the LSST Year 3 framework of Mitra et al. (2023), which built upon the Photometric LSST Astronomical Time-Series Classification Challenge (PLAsTiCC) framework"

    The forecast pipeline cites the author's own prior LSST Y3 mock framework. This is ordinary self-citation for projections only; it does not underwrite the paper's central late-time null result (Table 2), which uses independent public DES/DESI/RSD data. Not load-bearing, hence score remains 1 rather than higher.

full rationale

The paper's load-bearing claim is a null detection: joint DES-SN5YR (Dovekie) + DESI 2024 BAO (+ compact RSD) yield nearly identical chi^2_min to flat LambdaCDM (Table 2: 1656.15 vs 1656.07 background; 1664.18 vs 1663.95 joint), with Delta prior-dominated and information criteria/Bayesian evidence (dynesty log Z) penalizing extra parameters. This is obtained by MCMC sampling of an explicitly defined effective Omega_Lambda(z) against external public likelihoods; it is not forced by construction from the inputs. The tanh profile (Eq. 2) is introduced as a standard sigmoid smoothing of the prior step-function Lambda_sCDM, with normalization chosen only to enforce flatness at z=0 (a model definition, not a prediction). The analytic argument that Delta is unconstrained when z^dagger ~ 3.5 (matter domination, Omega_DE/Omega_m ~ 0.025) is a physical expectation confirmed by the posterior, not a circular reduction. Mild citations to the Lambda_sCDM literature (Akarsu et al.) and to the author's own prior LSST mock framework (Mitra et al. 2023) appear only for motivation and forecasts; they are not load-bearing for the late-time null. No fitted parameter is renamed a prediction, no uniqueness theorem is imported, and no self-definitional loop equates output to input. Score 1 only for the ordinary (non-load-bearing) self-citation in the forecast section.

Assumptions & free parameters 5 free parameters · 5 assumptions · 1 invented entities

The central null rests on standard FLRW cosmology plus a phenomenological tanh vacuum profile, free late-time calibration K, and a sub-horizon growth approximation. Free parameters are the usual cosmological densities plus the two transition parameters and sigma_8; the invented entity is the smooth sign-switching vacuum itself, which has no independent microphysical evidence in this work. No machine-checked formalization is claimed.

free parameters (5)
  • Omega_m0 = 0.3130^{+0.0132}_{-0.0114} (joint smooth model)
    Present-day matter density fitted freely in all MCMC runs; central to both background distances and growth.
  • K = c/(H0 rd) = 29.8333^{+0.2853}_{-0.2854} (joint)
    Unconstrained BAO calibration parameter that absorbs H0 and sound-horizon freedom; deliberately left free to isolate late-time geometry.
  • z^dagger = 3.5044^{+1.0319}_{-1.0603} (joint)
    Transition redshift where vacuum energy crosses zero; free parameter of the model under test.
  • Delta = 24.7536^{+17.4227}_{-17.4807} (joint)
    Transition smoothness; free parameter whose posterior is prior-dominated and is the focus of the null claim.
  • sigma_8 = 0.8157^{+0.0341}_{-0.0339} (joint smooth)
    Present-day amplitude of matter fluctuations, free only in the joint growth analysis.
assumptions (5)
  • domain assumption Flat FLRW Friedmann equation with effective Omega_Lambda(z) (Eq. 1).
    Standard cosmological background; spatial curvature is set to zero throughout.
  • ad hoc to paper Hyperbolic-tangent vacuum profile normalized so Omega_Lambda(0) = 1 - Omega_m0 - Omega_r0 (Eq. 2).
    Phenomenological smoothing chosen by the author; not derived from a microphysical potential in the main analysis.
  • domain assumption Sub-horizon linear growth ODE with matter-dominated initial conditions at x = -4 (Eq. 8).
    Standard approximation for f-sigma_8, but applied here despite w_eff divergence at the zero-crossing.
  • ad hoc to paper Seven RSD f-sigma_8 points treated as independent effective constraints.
    Explicit methodological choice in Sec. 3; full survey covariances are not used.
  • ad hoc to paper Flat priors on Omega_m0, K, z^dagger, Delta, sigma_8 as listed in Eqs. 10-14.
    Prior volume on Delta directly shapes the unconstrained posterior that underpins the null claim.
invented entities (1)
  • Smooth tanh sign-switching vacuum energy (Lambda_s CDM with finite Delta)
    purpose: Provide a continuous AdS-to-dS transition that can mimic late-time expansion effects associated with the neutrino-mass boundary tension without an instantaneous jump.
    The entity is phenomenological; the paper itself notes that a stable scalar-field embedding near the zero-crossing is an open theoretical problem (Appendix A).

how reviews work

0 comments
Cite this review

Pith. "Pith review of Catching the Cosmic Sign Flip: Background and Growth Tests of Smooth Sign Switching Lambda_s CDM." pith.science (2026). https://pith.science/paper/HWGNU7MX

@misc{pith2026260706735,
  author       = {Pith},
  title        = {Pith review of: Catching the Cosmic Sign Flip: Background and Growth Tests of Smooth Sign Switching Lambda_s CDM},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/HWGNU7MX}},
  note         = {Machine review of arXiv:2607.06735}
}
abstract

Recent baryon acoustic oscillation (BAO) measurements from the Dark Energy Spectroscopic Instrument (DESI), in combination with CMB and supernova data, show a mild preference for dynamical dark energy over flat $\Lambda$CDM. They can also drive the best-fit effective neutrino mass to unphysical negative values, motivating models that mimic this effect through late-time expansion physics. The sign-switching cosmological constant ($\Lambda_s$CDM) model addresses this by introducing an AdS-like negative vacuum energy density at high redshifts ($z > z^\dagger$) that transitions to a dS-like positive density at low redshifts. While original fits assumed a discontinuous step function transition, we generalise the dynamics using a smooth hyperbolic tangent parametrisation, $\Lambda(z) \propto \tanh[\Delta(z^\dagger - z)]$. We constrain the transition redshift $z^\dagger$ and smoothness parameter $\Delta$ using background distance data from DES-SN5YR (with Dovekie recalibration) and DESI 2024 BAO. We further incorporate a compact redshift space distortion (RSD) $f\sigma_8$ compilation. The joint data do not prefer the smooth sign-switching model over flat $\Lambda$CDM, yielding nearly identical $\chi^2_{\text{min}}$ values, while information criteria penalise the additional parameters. The transition smoothness remains unconstrained and dominated by the prior volume ($\Delta = 24.75^{+17.42}_{-17.48}$), indicating that current data cannot distinguish between a sharp transition and a smooth dynamical crossover. This work does not directly constrain the physical sum of neutrino masses $\sum m_\nu$; rather, it shows that current late-time background and growth data alone cannot resolve the transition details. We discuss the implications of this null detection and present projections for next-generation surveys.

Figures

Figures reproduced from arXiv: 2607.06735 by the authors.

Figure 1
Figure 1. Joint posterior distributions (corner plots) (Foreman-Mackey 2016) for flat ΛCDM (left) and CPL (right). The contours show 68% and 95% confidence regions, and the black markers represent the maximum likelihood points. Alt text: Two corner plots comparing posterior constraints for flat ΛCDM and CPL. Each panel shows one- and two-dimensional marginalized distributions with confidence contours and maximum-likelihood ma… view at source ↗
Figure 2
Figure 2. Joint posterior distribution (corner plot) (Foreman-Mackey 2016) for the smooth sign switching Λ𝑠CDM model. The contours show 68% and 95% confidence regions for Ω𝑚0, K = 𝑐/(𝐻0𝑟𝑑 ), 𝑧 † , and Δ; the black markers represent the maximum likelihood point. Alt text: Corner plot of posterior constraints for the smooth sign switching Λ𝑠CDM model, showing marginalized distributions and confidence contours for the matter den… view at source ↗
Figure 3
Figure 3. Distance ratio residuals for the DESI 2024 BAO data vector compared to the best fitting predictions of flat ΛCDM (top left), CPL (top right), and smooth sign switching Λ𝑠CDM (bottom) models. The error bars represent the diagonal 1𝜎 uncertainties from the DESI covariance matrix. Alt text: Three residual plots showing DESI 2024 BAO distance measurements relative to best-fitting flat ΛCDM, CPL, and smooth sign switchin… view at source ↗
Figures from the paper (2 more)
Figure 5
Figure 5. Figure 5: Comparison of the 2D joint posterior constraints (68% and 95% confidence regions) on the transition redshift 𝑧 † and the transition smooth￾ness parameter Δ. The wide blue contours represent the current constraints obtained in this work using real DES-SN5YR and DESI 202…
Figure 4
Figure 4. Figure 4: Evolution of the effective dark energy density fraction ΩDE (𝑧) (top), the expansion rate ratio 𝐻 (𝑧)/𝐻LCDM (𝑧) (middle), and the linear growth factor ratio 𝐷(𝑧)/𝐷LCDM (𝑧) (bottom) as a function of redshift 𝑧. We compare the best fitting flat ΛCDM (dashed grey), CPL (d…

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

39 extracted references · 39 canonical work pages

  1. [1]

    Qu, F. J. and others , title =. ApJ , year =. doi:10.3847/1538-4357/acfe06 , archiveprefix =

  2. [2]

    Madhavacheril, M. S. and others , title =. ApJ , year =

  3. [3]

    and Kumar, S

    Akarsu, O. and Kumar, S. and. Phys. Rev. D , year =

  4. [4]

    and Di Valentino, E

    Akarsu, O. and Di Valentino, E. and Kumar, S. and Nunes, R. C. and Vazquez, J. A. and Yadav, A. , title =. ArXiv e-prints , year =

  5. [5]

    and others , title =

    Alam, S. and others , title =. MNRAS , year =

  6. [6]

    Benson, B. A. and others , title =. Proc. SPIE Int. Soc. Opt. Eng. , year =

  7. [7]

    ArXiv e-prints , year =

    Bouhmadi-L. ArXiv e-prints , year =

  8. [8]

    and others , title =

    Brout, D. and others , title =. ApJ , year =

Show all 39 references
  1. [9]

    Abazajian, K. N. and others , title =. ArXiv e-prints , year =

  2. [10]

    Abbott, T. M. C. and others , title =. Phys. Rev. D , year =

  3. [11]

    Adame, A. G. and others , title =. JCAP , year =. doi:10.1088/1475-7516/2025/02/021 , archiveprefix =

  4. [12]

    and others , title =

    Di Valentino, E. and others , title =. Class. Quantum Grav. , year =. doi:10.1088/1361-6382/ac086d , archiveprefix =

  5. [13]

    and Hogg, D

    Foreman-Mackey, D. and Hogg, D. W. and Lang, D. and Goodman, J. , title =. PASP , year =

  6. [14]

    , title =

    Foreman-Mackey, D. , title =. J. Open Source Softw. , year =

  7. [15]

    Freedman, W. L. and others , title =. ApJ , year =

  8. [16]

    and others , title =

    Guy, J. and others , title =. A&A , year =

  9. [17]

    ApJS , year =

    Hlo. ApJS , year =. doi:10.3847/1538-4365/accd6a , archiveprefix =

  10. [18]

    and Ross, A

    Howlett, C. and Ross, A. J. and Samushia, L. and Percival, W. J. and Manera, M. , title =. MNRAS , year =

  11. [19]

    and Kamionkowski, M

    Karwal, T. and Kamionkowski, M. , title =. Phys. Rev. D , year =

  12. [20]

    Kass, R. E. and Raftery, A. E. , title =. J. Am. Stat. Assoc. , year =

  13. [21]

    Kenworthy, W. D. and others , title =. ApJ , year =

  14. [22]

    and others , title =

    Kessler, R. and others , title =. PASP , year =

  15. [23]

    ArXiv e-prints , year =

    K. ArXiv e-prints , year =

  16. [24]

    Kumar, D. R. and Yadav, S. K. , title =. ArXiv e-prints , year =

  17. [25]

    , title =

    Lewis, A. , title =. JCAP , year =. doi:10.1088/1475-7516/2025/08/025 , archiveprefix =

  18. [26]

    and Kessler, R

    Mitra, A. and Kessler, R. and More, S. and Hlo. ApJ , year =. doi:10.3847/1538-4357/acb057 , archiveprefix =

  19. [27]

    and others , title =

    Aghanim, N. and others , title =. A&A , year =

  20. [28]

    and others , title =

    Aghanim, N. and others , title =. A&A , year =. doi:10.1051/0004-6361/201833910 , archiveprefix =

  21. [29]

    and others , title =

    Akrami, Y. and others , title =. A&A , year =

  22. [30]

    and others , title =

    Popovic, B. and others , title =. ArXiv e-prints , year =

  23. [31]

    and Smith, T

    Poulin, V. and Smith, T. L. and Karwal, T. and Kamionkowski, M. , title =. Phys. Rev. Lett. , year =

  24. [32]

    and others , title =

    Raghunathan, S. and others , title =. ArXiv e-prints , year =

  25. [33]

    Riess, A. G. and others , title =. ApJL , year =. doi:10.3847/2041-8213/ac5c5b , archiveprefix =

  26. [34]

    and Ye, G

    Rosatello, B. and Ye, G. and Berti, M. and Tutusaus, I. and Grimm, N. and Bonvin, C. , title =. ArXiv e-prints , year =

  27. [35]

    and others , title =

    Ade, P. and others , title =. JCAP , year =

  28. [36]

    Speagle, J. S. , title =. MNRAS , year =

  29. [37]

    ArXiv e-prints , year =

  30. [38]

    and Lewis, A

    Torrado, J. and Lewis, A. , title =. JCAP , year =

  31. [39]

    and others , title =

    Alam, S. and others , title =. Phys. Rev. D , year =

Pith tools

Reviewed July 10, 2026 · model on record in the stance chip above.