{"id":"16335d01-657f-4142-bcbc-072f43ec3fdc","arxiv_id":"2412.14139","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":5.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":3,"one_line_summary":"A one-parameter interacting dark sector model can shift H0 upward and S8 downward when fit to Planck, DESI and supernova data, but the Bayesian evidence over LambdaCDM is inconclusive without SH0ES calibration.","lead":"This paper fits a model in which dark matter and dark energy are two scalar fields that interact with each other to the latest cosmological datasets. The interaction can raise the inferred Hubble constant and slightly lower a cosmic structure parameter, but the improvement is modest and depends on which datasets are included.","discovery_kind":"new_application","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The Table II constraints and tension claims depend on perturbation equations that are neither derived in this paper nor released in code: Section II defers to [31] and Section III A uses an unreleased modified CLASS. This is the load-bearing unverified link.","rationale":"I have read the paper in good faith. The hybrid model is clearly formulated at the background level, the datasets are standard, and the authors are appropriately cautious: they report negative Bayesian evidence for most combinations, the residual H0 tension is still 4.65 sigma, and they include SDSS and A_L checks. The central claim, however, rests on the linear perturbation predictions of the model. Those are not derived in the paper and the implementation is not public, so the correctness of the constraints cannot be independently checked. That is the same weakest assumption identified by the reader, and it justifies the CONDITIONAL verdict. The prior-range issue (sampling 1/phi_i up to 1 when Eq. (7) bounds it at about 0.577) is real and should be fixed, especially for the Bayesian evidence numbers, but it does not change the main tension-alleviation results. Because my concern matches the reader's and does not push the verdict in a new direction, I recommend no change to the reader's CONDITIONAL verdict.","tokens_in":20567,"tokens_out":8476,"duration_ms":83024,"concrete_test":"Make the modified CLASS source and the MontePython likelihood configuration public, and independently rerun the Pl18+DESI+SH0ES case, checking that Table II reproduces: H0=70.30±0.56 km/s/Mpc, 1/phi_i=0.0570(+0.0096,-0.0070), Delta ln B=2.5. In parallel, derive the linear perturbation equations from the action in Eqs. (1)-(4) and compare them with [31], specifically the effective sound speed and momentum exchange for the oscillating chi field. If the public rerun differs by more than ~0.1 sigma, or if the independently derived perturbations disagree with [31], the reported constraints and significance claims are not reliable.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central numerical claims—the 2–3 sigma detection of 1/phi_i, the shift of H0 to about 70.3, the mild S8 reduction, and the quoted tension metrics—are outputs of a MontePython+CLASS pipeline whose model-specific physics is not present in this article. The background equations (9)–(10) are given, but the CMB and lensing likelihoods also require the perturbed dark-sector energy-momentum tensor: density and pressure perturbations, effective sound speed, and momentum exchange for the oscillating chi field. Section II states 'We refer to [31] for the complete derivation of the perturbation equations', and Section III A says 'we implement the relevant equations ... in our modified version of CLASS', but neither the equations nor the code are provided. If the sound speed or the coupling terms in those perturbation equations are wrong, or if the private CLASS implementation contains a bug, every posterior and every derived significance in Table II could shift. This is not a cosmetic reproducibility issue: the quantities that drive the model's appeal (higher H0, lower S8) are exactly the ones most sensitive to how extra pressure perturbations source the CMB gravitational potentials. A secondary but concrete issue is that the prior 1/phi_i in [0,1] includes values violating Eq. (7), whose validity bound is 1/phi_i < 1/sqrt(3) ≈ 0.577; this should be corrected in the evidence calculation, though it is less central than the perturbation-code gap.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper constrains a two-scalar-field 'hybrid' dark sector model, a one-parameter extension of ΛCDM in which the inverse initial value of the dark-energy field, 1/φ_i, sets the DM–DE coupling. Using Planck 2018 CMB data, DESI BAO, Pantheon+ supernovae (with and without SH0ES calibration), and robustness checks against SDSS BAO and an A_L extension, the authors report a 2σ detection of the coupling with DESI, a >3σ detection in SH0ES-inclusive combinations, an upward shift of H0 to about 70.3 km/s/Mpc, a mild S8 change, and a reduction of the Hubble tension from about 5.76σ to 4.65σ. Bayesian evidence is inconclusive for most combinations and moderate only when SH0ES is included. The analysis follows standard MCMC/profile-likelihood practice and includes convergence checks, a profile-versus-Bayesian comparison, and χ² breakdowns.","tokens_in":20944,"tokens_out":10795,"duration_ms":92556,"significance":"If the underlying perturbation treatment is correct, the paper presents a theoretically motivated model that can shift H0 and S8 simultaneously while improving the fit to DESI BAO relative to ΛCDM, with a testable prediction of a time-dependent dark-matter density. The study is careful in its use of standard tools (MontePython, CLASS, GetDist, Procoli, MCEvidence) and provides useful cross-checks, including an SDSS BAO comparison and an A_L robustness analysis. However, the central numerical results depend on perturbation equations and modified code that are not included in the manuscript, and one of the abstract-level claims (a mild decrease of S8) is not borne out by the reported constraints. These issues must be addressed before the paper's conclusions can be fully assessed.","major_comments":[{"comment":"The perturbation equations for the hybrid dark sector are not given. Section II states 'We refer to [31] for the complete derivation of the perturbation equations', and Section III A says 'we implement the relevant equations ... in our modified version of CLASS', but neither the equations nor the code are provided. The CMB and lensing likelihoods depend directly on the perturbed energy-momentum tensor of the oscillating χ field—its density and pressure perturbations, effective sound speed, and momentum exchange with φ—so the posterior on 1/φ_i and all derived quantities (H0, S8, tension metrics) rest on an unverifiable implementation. I ask that the full perturbation equations be included (e.g., in an appendix) or that the modified CLASS code be released with a version identifier, so that the results can be independently checked.","section":"II and III A"},{"comment":"The prior on 1/φ_i is listed in Table I as [0,1], but the model's validity condition in Eq. (7), 1 < (φ/M_Pl)^2/3, requires 1/φ_i < 1/√3 ≈ 0.577. The text in Section III A claims the uniform prior covers 'the range of validity of the model's assumptions', which is inconsistent with the stated range. Because the Bayesian evidence values in Table II are computed over this prior, the quoted log B values (e.g., 2.5 and 4.5) include a prior volume of unphysical parameter space. The evidence should be recomputed with a prior truncated to the validity bound (or the validity range should be re-derived and justified); this can shift log B by roughly 0.5, which may alter the strength of the reported evidence.","section":"Table I / Eq. (7)"},{"comment":"The abstract and Section III B state that the model yields a 'mild decrease of the weak-lensing parameter S8'. Comparing Table II (hybrid) with Table III (ΛCDM) for the same dataset combinations, the hybrid S8 is not consistently lower: for Pl18+DESI, S8 = 0.817±0.013 vs 0.810±0.012; for Pl18+DESI+SH0ES, S8 = 0.818±0.013 vs 0.794±0.011; and for Pl18+SH0ES, S8 = 0.809±0.014 vs 0.795±0.013. Only for Pl18 alone is the hybrid S8 lower. The text's statement in Section IV that 'the decrease in Ωm at late times dominates, yielding a slightly smaller S8' is not consistent with these numbers. Please clarify whether the claim refers to an internal correlation within the model rather than a reduction relative to ΛCDM at fixed data, and adjust the abstract and conclusions accordingly.","section":"Abstract / III B / Tables II and III"}],"minor_comments":[{"comment":"For the Pl18+DESI combination, the text reports a detection at '2 sigma' using a 95% CL interval whose lower bound is 0.004, which is just above zero; for Pl18+DESI+SN, the text says 'only at 1 sigma' although the reported 68% interval is 0.029+0.017−0.015, with lower bound 0.014 > 0. Please define the significance convention used (e.g., 68%, 95%, 99% CL) for each detection claim, so the statements are unambiguous.","section":"III B, Table II"},{"comment":"The discussion of the redshift-dependent DM density compares the hybrid model to ΛCDM with 'the same best-fit parameters' in the top panel, but it would help to state explicitly in the caption that the yellow curve uses the hybrid best-fit cosmological parameters with ΛCDM dynamics, while the grey curve uses the ΛCDM best fit; this distinction is clear in the text but not in the caption.","section":"III B, Fig. 5"},{"comment":"The paper links the public CLASS and MontePython repositories but not the modified CLASS version used for the hybrid model. Even if the perturbation equations are added to the paper, a public link (or a clear statement on availability upon request) would improve reproducibility.","section":"II and III A"}],"recommendation":"major_revision","confidential_remarks":"The main barrier to verification is the unreleased modified CLASS code and the missing perturbation equations. The journal may wish to make code release a condition for publication. The prior-range inconsistency (Table I vs Eq. (7)) suggests the authors should re-check how the prior was implemented, as it affects the Bayesian evidence values that support the 'moderate evidence' claim. The S8 statement in the abstract appears to overstate what the constraints show, so I recommend the authors revisit it carefully before resubmission."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"The short version: this paper constrains a one-parameter interacting dark sector model (two scalar fields, DM and DE) against Planck 2018, DESI BAO, Pantheon+ and SH0ES. The model itself was proposed in 2023 by three of the same authors; what is new here is the first full MCMC treatment and the claim that DESI prefers a non-zero DM-DE coupling at about 2 sigma, and more than 3 sigma when SH0ES is added. The Hubble tension drops from about 5.8 to 4.7 sigma, which is real but partial, and the S8 shift is mild. The authors are careful not to oversell: Bayesian evidence is moderate only in the SH0ES-inclusive combinations, and they give chi-square breakdowns and a robustness check with SDSS BAO.\n\nWhat the paper does well: the analysis is standard and transparent. They use CLASS and MontePython, check convergence, run profile likelihoods, and report both delta-chi-squared and Bayesian evidence. The SDSS comparison is a good control, and the appendix on the lensing amplitude is honest. The effective dark energy equation of state never becomes phantom, which is a useful contribution to the DESI w0-wa discussion.\n\nThe soft spots. The main one is that the perturbation equations are not derived here (deferred to ref [31]) and the modified CLASS code is not released. For a paper whose central result is a detection of the coupling from CMB data, this is a real reproducibility gap. If the effective sound speed or momentum exchange terms are wrong, the posteriors could shift. I would not call it fatal, since the background equations are given and the model paper is by the same group, but the authors should be asked to put the perturbed equations in an appendix and release the code. A smaller issue: the prior on 1/phi_i goes from 0 to 1, but the model validity bound from Eq. (7) is 1/phi_i < about 0.577. The posterior peaks at about 0.06, so the constraints are not affected, but the Bayesian evidence is computed over a partly invalid region, which is sloppy for a model comparison claim.\n\nBottom line: this is a serious paper that deserves a referee. The main claims are supported by standard tools and the caveats are appropriate. I would send it to review, with a request to address the perturbation equation reproducibility and the prior range.","headline":"First serious constraints on a hybrid dark sector model; the DESI preference for a non-zero coupling is interesting, but the perturbation equations live in an unreleased code.","tokens_in":21499,"tokens_out":3283,"would_cite":true,"duration_ms":30088,"reading_group":"yes","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":[],"pacs":[],"model":"deepseek-v4-flash","headline":"A one-parameter hybrid dark sector, constrained by Planck 2018, DESI BAO, and SH0ES data, is detected at more than 3σ and reduces the Hubble tension from about 5.76σ to 4.65σ.","keywords":["hybrid dark sector","dark matter-dark energy interaction","Hubble tension","S8 tension","DESI baryon acoustic oscillations","scalar field dark energy","Bayesian model comparison"],"falsifier":"A combined measurement of the growth rate $f\\sigma_8(z)$ at $z \\approx 0.5$--$1$ from galaxy clustering and weak lensing would decide: the hybrid best-fit ($1/\\phi_i \\approx 0.057$) predicts a dark-matter density about $2.5\\%$ below $\\Lambda$CDM at late times (Fig. 5), altering $f\\sigma_8$ by a comparable amount; a growth history matching $\\Lambda$CDM to better than that deviation would falsify the detected coupling.","tokens_in":20393,"feed_emoji":"🌌","tokens_out":24836,"duration_ms":180771,"temperature":0.7,"pith_summary":"This paper argues that a single new parameter — the initial value of a scalar field that sets the strength of a dark-matter–dark-energy interaction — can shift the inferred expansion rate upward and slightly lower the fluctuation amplitude $S_8$, easing both the Hubble and $S_8$ tensions. When constrained with Planck 2018, DESI BAO, and Pantheon+ supernova data, the coupling $1/\\phi_i$ is detected at $2\\sigma$ from DESI and above $3\\sigma$ once the SH0ES Cepheid calibration is added. The Hubble tension falls from about $5.76\\sigma$ to $4.65\\sigma$, and Bayesian evidence moderately favours the hybrid model only when SH0ES is included. If the detection holds, the model offers a non-phantom mechanism for the late-time dynamics preferred by DESI data.","feed_headline":"Eases Hubble tension to 4.65σ with a hybrid dark sector","feed_subtitle":"One extra parameter moves H₀ to ~70 km/s/Mpc; DESI and SH0ES data prefer it over ΛCDM.","key_machinery":"The load-bearing object is the averaged two-fluid system for the oscillating $\\chi$ field and the slowly rolling $\\phi$ field: $\\dot{\\rho}_c + 3H\\rho_c = (\\dot{\\phi}/\\phi)\\,\\rho_c$ and $\\ddot{\\phi} + 3H\\dot{\\phi} = -(1/\\phi)\\,\\rho_c$. These equations, derived in the companion paper [31], encode the entire new physics: the dark-matter energy density decays faster than $a^{-3}$ as $\\phi$ begins to roll, while $\\phi$ itself tracks matter and later freezes as an effective cosmological constant. Because the coupling constant $g$ cancels from these equations, the initial condition $1/\\phi_i$ alone sets the deviation from $\\Lambda$CDM. The same equations are extended to linear perturbations (from [31]) and implemented in a modified Einstein–Boltzmann solver to compute CMB and matter power spectra.","core_discovery":"The central claim is that the hybrid dark sector — two coupled scalar fields, $\\phi$ as dark energy and $\\chi$ as dark matter, with potential $V(\\phi,\\chi) = V_0 + \\tfrac{1}{2} g^2 \\phi^2 \\chi^2$ — is a one-parameter extension of $\\Lambda$CDM that improves the description of the combined Planck, DESI, and Pantheon+ data. The only new parameter is the initial value of the dark-energy field, $1/\\phi_i$, which sets the strength of the energy transfer from dark matter to dark energy. With best-fit values in the range $1/\\phi_i \\approx 0.03$--$0.06$, the model raises $H_0$ to about $70$ km/s/Mpc, lowers the physical dark-matter density $\\omega_c$, and produces a mild decrease of $S_8$ via the correlation between the coupling and $\\Omega_m$. The paper quantifies the detection at $2\\sigma$ with DESI and more than $3\\sigma$ with SH0ES, reports a reduction of the Hubble tension from $Q \\approx 5.76\\sigma$ to $\\approx 4.65\\sigma$, and finds moderate Bayesian evidence ($\\ln B \\approx 2.5$) only when the SH0ES calibration is included.","pith_inferences":["If the SH0ES-driven detection of $1/\\phi_i$ survives future data, the hybrid model would provide a non-phantom explanation for DESI's preference for dynamical dark energy, distinguishable from the CPL parametrisation by the shape of the growth history and the absence of $w < -1$.","Because the coupling constant $g$ drops out of the fluid equations, a confirmed nonzero $1/\\phi_i$ would leave the dark-matter mass $m_\\chi = g\\phi$ unconstrained by background data; translating the model into a particle-physics target (e.g. $g \\lesssim 10^{-8}$ from the non-oscillation condition) would require direct or indirect detection of the field's mass.","The paper restricts to adiabatic initial conditions; allowing isocurvature modes, which it flags as future work, could either weaken or sharpen the claimed detection, and should be a priority for next-generation CMB and LSS analyses.","A clean test of the mechanism is to measure the matter density as a function of redshift: the model predicts $\\rho_c$ is about $2.5\\%$ larger than $\\Lambda$CDM at $z \\gtrsim 1$ and $2.5\\%$ smaller today, a distinctive signature that lensing and galaxy-clustering surveys can target directly."],"forward_implications":["With Planck 2018 alone, the hybrid model is indistinguishable from $\\Lambda$CDM and yields the upper bound $1/\\phi_i < 0.039$.","Adding DESI BAO moves the coupling away from zero at $2\\sigma$, improves the fit by $\\Delta\\chi^2_{\\rm min} = -2.8$ relative to $\\Lambda$CDM, but leaves the Bayesian evidence inconclusive.","Including the SH0ES Cepheid calibration produces a detection at more than $3\\sigma$ ($1/\\phi_i = 0.057$) and moderate evidence in favour of the hybrid model ($\\ln B = 2.5$), while worsening the DESI fit.","The coupling $1/\\phi_i$ correlates positively with $H_0$ and negatively with $\\omega_c$ and $S_8$, allowing the model to raise the expansion rate while nudging $S_8$ downward; the $H_0$ tension metric drops from $5.76\\sigma$ to $4.65\\sigma$.","The model predicts a redshift-dependent dark-matter density — about $2.5\\%$ below $\\Lambda$CDM today at best fit — and an effective dark-energy equation of state that never crosses the phantom divide ($w \\ge -1$)."],"supporting_citations":[{"why":"This reference supplies the hybrid model and the averaged fluid and perturbation equations that the entire analysis implements.","marker":"[31]"},{"why":"This reference provides the Planck 2018 CMB, lensing, and cosmological baseline used in every dataset combination.","marker":"[12]"},{"why":"This reference provides the DESI Y1 BAO measurements that drive the $2\\sigma$ detection of the coupling.","marker":"[32]"},{"why":"This reference provides the SH0ES Cepheid distance calibration that turns the detection into a $>3\\sigma$ result and yields moderate Bayesian evidence.","marker":"[9]"},{"why":"These references supply the Pantheon+ supernova distance moduli used with and without the SH0ES calibration.","marker":"[58, 59]"},{"why":"These references describe the Einstein–Boltzmann solver whose modified version computes the CMB and matter power spectra for the hybrid model.","marker":"[44–46]"},{"why":"These references provide the Monte Carlo sampler interfaced with the modified solver to generate the posterior chains.","marker":"[47, 48]"},{"why":"This reference supplies the profile-likelihood minimizer used for the $\\Delta\\chi^2$ differences and the DMAP tension metric.","marker":"[61]"},{"why":"These references provide the Bayesian evidence calculation used for the model comparison log B.","marker":"[62, 63]"}],"fun_headline_variants":["Hybrid dark sector cuts Hubble tension to 4.65σ","One-parameter model eases Hubble tension, hints at S8","Two-field dark sector nudges H0 up, S8 down","Dark coupling model reduces Hubble tension, Bayesian win with SH0ES"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The constraints assume that the averaged fluid equations and the linear perturbation equations for the oscillating $\\chi$ field, taken from the companion paper [31] and implemented in a modified Einstein–Boltzmann solver, correctly describe the dark sector; if those equations or their code implementation are wrong, the inferred posterior on $1/\\phi_i$ and all derived shifts in $H_0$ and $S_8$ would not be reliable.","fun_headline_variants_meta":{"raw":{"variants":["Hybrid dark sector cuts Hubble tension to 4.65σ","One-parameter model eases Hubble tension, hints at S8","Two-field dark sector nudges H0 up, S8 down","Dark coupling model reduces Hubble tension, Bayesian win with SH0ES"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000522,"raw_usage":{"total_tokens":2588,"prompt_tokens":1074,"completion_tokens":1514,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":690,"completion_tokens_details":{"reasoning_tokens":1439}},"tokens_in":690,"tokens_out":1514,"duration_ms":11096,"temperature":1.0,"reasoning_tokens":1439,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-11T12:26:55.486301+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"A combined measurement of the growth rate $f\\sigma_8(z)$ at $z \\approx 0.5$--$1$ from galaxy clustering and weak lensing would decide: the hybrid best-fit ($1/\\phi_i \\approx 0.057$) predicts a dark-matter density about $2.5\\%$ below $\\Lambda$CDM at late times (Fig. 5), altering $f\\sigma_8$ by a comparable amount; a growth history matching $\\Lambda$CDM to better than that deviation would falsify the detected coupling.","supporting_citations":[],"review_version":1}