Distinguishing Coupled Dark Matter Dark Energy from Kinematic Phantom Crossing

The theoretical spine is coherent: the coupling β enters the background (Eq. 13), the running CDM density (Eq. 3), the fifth force (Eq. 15), the drag −βφ′ (Eq. 16) and the non-conservation rate ε≡βφ′=dln f/dN consistently, and the φ_0≡1 gauge is used uniformly (Eq. 23 follows correctly from Eq. 8). However, there are multiple numerical mismatches between the narrative text and the tables that a careful reader cannot reconcile. (1) Sec. IV C states √Δχ²=20.8 'for the DES-SN5YR posterior median with the exponential potential, and 20.8–25.8 across all six chains', while Table III gives 22.5 for that exact entry and a range 20.3–25.6; my own check (Δr_s/r_s=0.88% ⇒ |Δℓ_A|≈2.65, /0.117≈22.7) supports the table, not the text. (2) The same paragraph quotes ω_c(z*)=0.1235–0.1257 against Table III's 0.1234–0.1254, and Ω_φ(z*)=4.9–6.4×10^-4 against Table III's 4.5–6.1×10^-4; the abstract's '4.4%–5.5%' mass change against Table III's 4.32%–5.49%. (3) The drag-to-fifth-force ratio is quoted three different ways without reconciliation: −0.603 to −0.635 from the integrator, r=−0.540±0.020 from the K-fit, and −0.66 from the matter-era estimate; Eq. (30)'s growth coefficient 2.24 uses c_5(1+r) with r=−0.540, yet Table IV's own C_5th+C_drag for DES/exp is 0.1724 versus 0.219 predicted by that term — a ~27% internal disagreement in the decomposition that is masked only because the total C_rel still lands within 3%. (4) Cross-reference slips: 'Table VI repeats Table I' (it repeats Table II); the α²>0.05 and α²>0.10 slopes are attributed to Table IV, which contains neither; the Introduction's section map (Sec. III = results) disagrees with the actual numbering (Sec. IV = results); and the Conclusion's opening sentence states the opposite of the paper's thesis ('whether a phantom-like crossing ... can originate from a fundamental phantom degree of freedom'). Individually these are wording/bookkeeping issues, but collectively they suggest the prose was written against a slightly different chain than the tables, and they affect secondary quantitative conclusions (the size of the recombination bias and the three-channel split) rather than the core argument, which survives. Panel split 2, 3, 3, 5 across 4 math specialists. The displayed score follows the conservative panel anchor.

The formal background mechanism is largely sound: Eqs. (3)–(7) consistently show how a canonical scalar plus energy exchange can yield an effective w below −1, and the perturbative eigenvalue calculation leading to p_g=1+(6/5)f_c²β²+O(β⁴) is plausibly reproducible from Eq. (25). Equation (27) is also appropriately labeled as a fifth-force-only matter-era estimate. However, the central numerical inference is not mathematically validated for the stated model. The paper uses a ΛCDM-trained CAMB emulator, substitutes a single equality-era CDM density into its transfer function, and applies compressed CMB priors derived under a different evolution law. No coupled Boltzmann derivation or quantitative error propagation establishes that these approximations are accurate enough for the reported percent-level slopes and log-evidence differences. The initial conditions in Eq. (18) also do not match the coupled two-fluid growing mode derived later, and the definition of δ_m in Eq. (19) is not reconciled with running species weights or massive neutrinos. Equation (30) is chiefly a numerical regression with fitted exponents, not a derivation from dimensional analysis. Since these gaps affect the main growth and inference claims, a score above 3 is excluded; their combination warrants 2. Panel split 2, 2, 4, 4 across 4 math specialists. The displayed score follows the conservative panel anchor.

Empirical rubric applied. The paper offers multiple specific, quantitative, and jointly falsifiable predictions: a P m_nu ∝ f_c^2 alpha^2 degeneracy scaling with explicit numerical coefficients (Eq. 30), a 4.4%-5.5% CDM mass increase at recombination with concrete ω_c(z*) values (Table III), a predicted crossing redshift range z_×=0.66-0.75 distinct from w0waCDM's z_×=0.35-0.43, and a drag-to-fifth-force cancellation ratio of approximately -0.54 to -0.64. These are tested against current DESI DR2, ACT DR6, Planck-derived, and RSD data with quoted posteriors and Bayes factors, and the authors explicitly note upcoming Euclid-era growth/lensing data as a further test. This is not operationally empty: current data already partially constrain these predictions (Bayes factors 2-17, moderate to strong on Jeffreys scale). The score is capped at 4 rather than 5 because the coupling detection is only moderate-strength (not decisive), the discriminating power between CDEDM and phenomenological models rests substantially on Bayesian model comparison and a specific decomposition that has not yet been independently confirmed by another team, and the recombination-era treatment is explicitly flagged by the authors as an approximation pending a full coupled-Boltzmann calculation.

The paper has a strong overall organization, defines the physical mechanism early, presents data choices and limitations explicitly, and separates analytic scaling arguments from numerical calibration. A graduate cosmology reader can follow its central narrative. However, the communication is weakened by an overextended results discussion, repeated claims that the model can 'distinguish' scenarios before a full observable-level discrimination analysis is supplied, and a tension between the headline f_c^2 alpha^2 scaling and the later result that the total relation is dominated by a shooting-dependent background contribution with fitted exponents. The explicit dual use of lambda is flagged, so it does not trigger the symbol-redefinition cap, but it remains a readability burden.

Coupled quintessence, effective phantom crossing, fifth-force growth enhancement, and coupling–neutrino-mass degeneracies are established research directions and are appropriately cited. The potentially new contribution is the specific decomposition of the inferred degeneracy into background, fifth-force, and drag channels; the numerical calibration of their relative contributions; and the emphasis on correlated recombination-era running-mass effects. This is a meaningful synthesis with potentially useful discriminants, but the paper itself shows that the total fitted relation is substantially background-driven and potential-dependent, rather than a wholly new universal f_c^2 alpha^2 law. Its novelty therefore lies more in model-specific quantitative organization and calibration than in a new underlying mechanism.

The core scientific argument is well-structured: assumptions (canonical scalar, Yukawa coupling, uncoupled baryons) are stated up front, the background and perturbation equations are derived with defined variables, boundary/initial conditions are specified (shooting procedure, quasi-static attractor, Appendix A-B), and limitations are repeatedly and explicitly acknowledged (approximate transfer function, no full coupled-Boltzmann code, non-closed-form background contribution, λ-range limitations for the power-law potential, prior-boundary caveats for Bayes factors). The stated goal of finding growth/recombination signatures that break the phantom-crossing degeneracy is substantively pursued through multiple independent channels (degeneracy scaling, mass running, energy-exchange rate, tension implications) with quantitative decomposition (Table IV) and robustness checks (Appendix D-E, prior sensitivity). However, completeness is meaningfully undercut by reference-list integrity problems: the automated verification found 25 fabricated citations (identifiers resolving to nothing with no matching underlying work found) among 92 checked, plus a further 28 unverified. Several of these fabricated citations support claims used to justify methodological choices and background physics (e.g., BAO/CMB references, coupled-quintessence foundational citations), which is a serious completeness/rigor concern even though it does not break the internal mathematical logic of the paper itself. This citation problem, combined with some acknowledged but unresolved approximations in the recombination/transfer-function treatment, keeps the score at a mid-level rather than the top of the scale.