{"id":"425fc2e2-c751-4e8a-bc9d-91cd6c529b9f","arxiv_id":"2607.05044","paper_version":1,"verdict":"CONDITIONAL","confidence":"HIGH","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":3,"one_line_summary":"Smooth ECDM dark energy remains compatible with Planck+ACT+SPT, DESI DR2 and Pantheon+/SH0ES while alleviating the Hubble tension through a controlled late-time density transition.","lead":"A smooth sign-switching dark energy model (ECDM) fits full CMB, DESI BAO and Pantheon+ data while raising H0 and easing the Hubble tension. The work supplies regular perturbation equations that stay well-behaved when the dark-energy equation of state diverges at the sign change.","discovery_kind":"extension","skeptic_critique":{"model":"grok-4.5","headline":"The statistical preference and H0 relief rest on a prior-truncated upper bound for η that the paper itself flags as a fluid-model breakdown, so the claimed compatibility may be an artifact of the regularised-perturbation formulation.","rationale":"The paper is technically careful: the reformulation of the continuity and Euler equations in terms of fA is well-motivated, the multi-experiment CMB likelihoods are state-of-the-art, and the power-spectrum residuals are cleanly presented. The reader’s weakest-assumption diagnosis (fixed cs^{2}=1 plus the two-parameter erf profile) is correct but secondary. The more immediate load-bearing issue is internal to the perturbation scheme the authors introduce: the same regularisation that lets them evolve through ρd=0 produces a data-independent upper bound on η that excludes the fast-transition regime previously shown (and reconfirmed here with fixed η) to give the largest H0 relief. Because the abstract and conclusions present ECDM as “fully compatible … while alleviating the Hubble tension,” this circularity between the numerical method and the statistical preference is the single most important caveat. The verdict therefore remains CONDITIONAL, but the concrete stress-test should target the fluid-to-field replacement rather than merely varying cs^{2}. No evidence of fraud or over-claim beyond ordinary phenomenological modelling is present; the concern is purely technical.","tokens_in":40873,"tokens_out":799,"duration_ms":7208,"concrete_test":"Re-run the CMB+PPS+DESI MCMC of Table 2 after replacing the fluid module with a scalar-field realisation of the same erf density profile (or with a hard cut-off that freezes DE perturbations for |x-x†|<1/η). If the upper bound on η disappears and the MAP H0 moves above ~71 km s^{-1} Mpc^{-1} without a 10 % χ^{2} penalty, the fluid-regularisation artifact is confirmed and the headline claim weakens.","verdict_should_be":"CONDITIONAL","load_bearing_attack":"The central claim (abstract, §6–8) that ECDM is fully compatible with the full CMB+DESI+Pantheon+/SH0ES suite while alleviating the Hubble tension is load-bearing on the posterior for the transition speed η. Table 2 and Fig. 7 show that every data combination returns only an upper bound on log10 η (or η itself). Section 6.2 explicitly states that this bound is data-independent and originates from the breakdown of the dark-energy fluid description: after the sign switch the rescaled density contrast fd develops oscillatory behaviour (visible in the lower-centre panel of Fig. 1 and in the fixed-η=10^{3/2} experiment of Fig. 8). When the authors force the largest prior value η=10^{3/2}, H0 rises toward the SH0ES value but S8 jumps to ~0.90 and the best-fit χ^{2} degrades by ~10 %. Thus the region of parameter space that most effectively relieves the Hubble tension is precisely the region the regularised fluid equations themselves exclude. The reader correctly flags the fixed cs^{2}=1 choice, but the more immediate threat is that the very variables introduced to keep the equations regular (fA=δA/(1+wA), Eqs. 4.4–4.5) generate the artificial upper bound that shapes the entire statistical preference.","agreement_with_reader":"partial"},"referee_report":{"model":"grok-4.5","summary":"The paper introduces a consistent linear-perturbation formulation for the ECDM model (error-function sign-switching dark energy density, Eq. 3.1) that remains regular when w_d diverges at the zero-crossing of ρ_d. Using a modified CAMB + Cobaya MCMC pipeline, the authors confront the model with Planck 2018 + ACT DR6 + SPT-3G CMB (including lensing), DESI DR2 BAO, Pantheon+ SNe, and SH0ES H0. They report that ECDM is compatible with the full data suite, yields a mild upward shift in H0 relative to ΛCDM (Table 2), improves most information criteria and tension metrics when SH0ES is included, and produces distinctive but observationally acceptable signatures in the matter power spectrum, ISW, and CMB lensing (Figs. 9–11).","tokens_in":41284,"tokens_out":1410,"duration_ms":22139,"significance":"If the results hold, the work supplies the first full-CMB, perturbation-consistent test of a smooth sign-switching DE scenario, closing a technical gap left by earlier background-only or abrupt-transition (ΛsCDM) studies. The regularised variables f_A ≡ δ_A/(1+w_A) (Eqs. 4.4–4.5) and the accompanying synchronous-gauge implementation are reusable for any DE model that crosses ρ=0 or w=−1. The multi-probe power-spectrum analysis and the extensive suite of AIC/BIC/DIC/WAIC/evidence/suspiciousness diagnostics further strengthen the empirical case that late-time sign-switching remains viable. These elements constitute a clear incremental advance for the sign-switching literature and for the broader CosmoVerse programme.","major_comments":[{"comment":"§6.1–6.2 and Table 2: every data combination returns only an upper bound on log10 η (or η). The text itself states that this bound is “data-independent” and “originates from the breakdown of the dark-energy fluid description” (oscillatory f_d after the transition, lower-centre panel of Fig. 1 and the fixed-η=10^{3/2} experiment of Fig. 8). When the largest prior value is forced, H0 rises toward the SH0ES value but S8 jumps to ~0.90 and χ² degrades by ~10 %. Thus the region that most effectively alleviates the Hubble tension is precisely the region excluded by the regularised fluid equations. The central claim of “full compatibility while alleviating the Hubble tension” therefore rests on a prior-truncated, model-breakdown bound that must be either (i) shown to be physical rather than an artefact of the perfect-fluid closure, or (ii) replaced by a more fundamental (e.g. scalar-field) real","section":"§6.1–6.2, Table 2, Figs. 1 & 8"},{"comment":"§4.1: the rest-frame sound speed is fixed by hand to c_s^{2}=1 “for simplicity,” with only a brief check that small positive values leave the spectra unchanged. Because c_s^{2} directly controls the DE clustering that sources the ISW and the late-time growth features used to disfavour rapid transitions, the statistical preference for ECDM (and the upper bound on η) could shift under a free or scale-dependent c_s^{2}. A short MCMC exploration of c_s^{2} (or at least a clear demonstration that the posterior on {η,z†,H0} is insensitive) is required for the load-bearing claims of §§6–8.","section":"§4.1"},{"comment":"CMB-only posteriors are bimodal (Fig. 2), with a slow-transition branch that yields H0≳80 km s^{-1} Mpc^{-1} but is later eliminated by SNe. The paper reports the combined-data H0≃69 km s^{-1} Mpc^{-1} as “alleviating” the tension, yet the quantitative reduction relative to ΛCDM is modest (~0.8 km s^{-1} Mpc^{-1}) once the slow branch is removed. A clearer statement of the residual tension (in σ) for the fast branch alone, both with and without SH0ES, would prevent over-statement of the model’s success.","section":"§6.1, Fig. 2, Table 2"}],"minor_comments":[{"comment":"Title and abstract contain typographical artefacts (“Hubble T ension”, “PantheonPlus”). Standardise to “Hubble Tension” and “Pantheon+” throughout.","section":"title, abstract"},{"comment":"Eq. (3.2) and the subsequent total-w expression (3.3) would benefit from an explicit statement that the apparent pole in w_d is integrable and does not affect the background expansion; a short analytic check would help non-specialist readers.","section":"§3"},{"comment":"Figure 1 caption refers to “MAP values form the CMB-SPA combination”; correct the typo and define “SPA” (or replace by the explicit data combination used).","section":"Fig. 1"},{"comment":"The prior table (Table 1) lists both log10(log10(1+z†)) and the linear z†; clarify which parametrisation is actually sampled and which is derived.","section":"Table 1"},{"comment":"Appendix C defines several Bayesian estimators but does not state the precise value of α used for the truncated harmonic-mean evidence; the text later adopts α=0.95—move that choice into the appendix for reproducibility.","section":"Appendix C"}],"recommendation":"major_revision","confidential_remarks":"The manuscript is a solid, technically careful extension of the authors’ own background-only ECDM papers. The main risk is that the fluid-regularisation artefact on η is under-emphasised relative to the strong language of the abstract and conclusions. Once the three major points are addressed (or the claims are suitably caveated), the paper will be a useful addition to the sign-switching literature and appropriate for JCAP. No concerns about novelty disclosure or citation pattern."},"author_rebuttal":null,"desk_editor":{"model":"grok-4.5","letter":"The real advance here is the regularised perturbation system (f_A = δ_A/(1+w_A)) that stays finite when ρ_d crosses zero, its CAMB implementation, and the first joint Planck+ACT+SPT+DESI+Pantheon+/SH0ES MCMC that includes growth and lensing. Background-only ECDM already existed; this is the first time the model is properly tested against full CMB spectra.\n\nThey do the technical work carefully. Newtonian- and synchronous-gauge equations are derived cleanly, the MCMC is standard Cobaya with R–1 < 0.02, multiple chains, and a full suite of AIC/BIC/WAIC/DIC/evidence and tension metrics. Posteriors, power-spectrum residuals, and the ISW/lensing discussion are transparent. When the full data combination is used, the slow-transition branch is killed, H0 moves to ~69 km/s/Mpc, and most information criteria prefer ECDM over ΛCDM (especially with SH0ES). That is a genuine, usable result for the sign-switching literature.\n\nThe soft spot is real but not hidden. Every data combination returns only an upper bound on η; §6.2 states that the bound is data-independent and comes from the fluid description itself (fd starts oscillating after the switch). Forcing the prior edge η = 10^{3/2} pushes H0 toward SH0ES but drives S8 ~0.90 and worsens χ^{2} by ~10 %. So the region that most efficiently relieves the Hubble tension is exactly the region the regularised equations exclude. Fixed c_s^{2} = 1 and the pure error-function profile are ordinary phenomenological choices; they matter less than this self-imposed cutoff. No public code is a minor practical annoyance.\n\nThis is for people already working on late-time DE extensions or the H0 tension. It is not a microphysical model and does not close the tension, but it is a clean, honest step that deserves a serious referee. I would engage with it and expect it to be cited in the next round of sign-switching papers.","headline":"Solid first full-perturbation constraints on smooth sign-switching DE; the H0 relief and statistical preference are real but sit on a fluid-model upper bound for transition speed that the authors themselves flag.","tokens_in":41925,"tokens_out":563,"would_cite":true,"duration_ms":6148,"reading_group":"yes","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":[],"pacs":[],"model":"grok-4.5","headline":"Smooth sign-switching dark energy fits full CMB, DESI and Pantheon+ data while easing the Hubble tension.","keywords":["sign-switching dark energy","ECDM","Hubble tension","cosmological perturbations","CMB","DESI BAO","Pantheon+","ISW effect"],"falsifier":"A decisive increase in the best-fit χ^{2} when the transition speed is forced to extreme values, or a statistically significant mismatch between the predicted fσ_8(z) or CMB-lensing spectrum and forthcoming redshift-space-distortion or lensing data.","tokens_in":41724,"feed_emoji":"⭐","tokens_out":786,"duration_ms":6310,"temperature":0.7,"pith_summary":"The paper tests a smooth version of sign-switching dark energy, called ECDM, in which the dark-energy density gradually rises from a negative value to a positive one. The authors rewrite the linear-perturbation equations so they stay finite even when the dark-energy equation of state diverges at the zero-crossing. They then fit the model to Planck, ACT and SPT CMB data, DESI BAO, Pantheon+ supernovae and the SH0ES Hubble-constant measurement. The joint analysis shows that ECDM remains compatible with all these probes, prefers a finite transition speed rather than an instantaneous jump, and moves the inferred Hubble constant closer to the local value, thereby reducing the long-standing tension with early-universe inferences.","feed_headline":"Smooth dark-energy switch eases Hubble tension","feed_subtitle":"Full CMB, DESI and Pantheon+ data still fit when density flips from negative to positive","key_machinery":"The rescaled density contrast f_A = δ\rho_A / (ρ_A + p_A), which remains finite when the dark-energy density crosses zero and thereby allows stable numerical evolution of perturbations through the sign switch.","core_discovery":"When linear dark-energy perturbations are treated consistently, the smooth error-function ECDM model is fully compatible with current precision CMB, BAO and supernova data and simultaneously lowers the Hubble tension relative to flat ΛCDM.","pith_inferences":["A microphysical scalar-field realisation would be needed to decide whether the upper bound on transition speed is generic or an artefact of the error-function ansatz.","The same regularisation of perturbations could be applied to other models that cross the null-energy condition or the zero-density line.","If the mild S_8 upshift survives, weak-lensing surveys may become the decisive arbiter between ECDM and ΛCDM."],"forward_implications":["Joint CMB+BAO+SNe analyses will continue to favour a finite transition speed over an instantaneous AdS-to-dS jump.","The model predicts a temporary negative ISW–galaxy cross-correlation around the transition redshift that future surveys can search for.","Ultra-fast transitions are ruled out by the growth of structure once dark-energy perturbations are included.","Early-universe physics remains essentially unchanged, so the sound-horizon scale is left intact."],"fun_headline_variants":["Smooth ECDM sign-switch fits CMB DESI Pantheon+ and eases H0 tension","Consistent DE perturbations keep sign-switching model viable vs full data","Negative-to-positive DE density transition softens Hubble tension","Error-function ECDM reconciles precision cosmology and lowers H0 gap","Sign-switching dark energy remains compatible while cutting Hubble tension"],"cache_read_input_tokens":32896,"weakest_assumption_plain":"The rest-frame sound speed of dark energy is fixed by hand to the speed of light, and the background density is assumed to follow a pure two-parameter error-function profile.","fun_headline_variants_meta":{"raw":{"variants":["Smooth ECDM sign-switch fits CMB DESI Pantheon+ and eases H0 tension","Consistent DE perturbations keep sign-switching model viable vs full data","Negative-to-positive DE density transition softens Hubble tension","Error-function ECDM reconciles precision cosmology and lowers H0 gap","Sign-switching dark energy remains compatible while cutting Hubble tension"]},"model":"grok-4.5","effort":"low","cost_usd":0.005796,"raw_usage":{"total_tokens":1514,"prompt_tokens":726,"num_sources_used":0,"completion_tokens":96,"cost_in_usd_ticks":57960000,"prompt_tokens_details":{"text_tokens":726,"audio_tokens":0,"image_tokens":0,"cached_tokens":256},"completion_tokens_details":{"audio_tokens":0,"reasoning_tokens":692,"accepted_prediction_tokens":0,"rejected_prediction_tokens":0}},"tokens_in":726,"tokens_out":96,"duration_ms":5625,"temperature":1.0,"reasoning_tokens":692,"cache_read_input_tokens":256,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-07-11T09:39:39.460854+00:00","model_set":{"reader":"grok-4.5"},"falsifier":"A decisive increase in the best-fit χ^{2} when the transition speed is forced to extreme values, or a statistically significant mismatch between the predicted fσ_8(z) or CMB-lensing spectrum and forthcoming redshift-space-distortion or lensing data.","supporting_citations":[],"review_version":1}