{"id":"a0e6b32a-2309-4d44-884b-5bca6c7383ce","arxiv_id":"2607.16158","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":4.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":5,"one_line_summary":"Using CMB+BAO+cosmic chronometer+Pantheon+ data, f(R,T)=R+λT^ε gravity is constrained to ε=0.010^{+0.013}_{-0.021}, consistent with ΛCDM's ε=0.","lead":"This paper fits a modified gravity model, f(R,T)=R+λT^ε, to CMB, BAO, cosmic-chronometer, and supernova data. It finds ε≈0.010, consistent with ordinary dark energy (ε=0), so the modification is not preferred.","discovery_kind":"extension","skeptic_critique":{"model":"deepseek-v4-flash","headline":"CMB distance priors from wCDM are applied to f(R,T) without validation; this could bias the ε posterior.","rationale":"I read the paper as a straightforward multi-dataset fit of f(R,T)=R+λT^ε. The central claim is that ε is constrained to be near 0, consistent with ΛCDM. The authors include a new CMB term via distance priors. The distance-prior approximation is common, but it is only reliable when the fiducial model of the prior is close to the model being tested. Here the f(R,T) model modifies the background expansion in a way that is not captured by wCDM, so the compressed likelihood may be biased. The paper does not validate this. The BAO exclusions are also underspecified, but the CMB transfer is more fundamental because it affects the early-universe constraints. The posterior is driven by the combination of CMB, BAO, SNe, and CC; a shift in the CMB term could propagate to ε. The paper provides code, which is good, but that does not settle the prior-transfer question. I agree with the reader's weakest_assumption and recommend keeping the conditional verdict pending self-consistent CMB likelihood checks.","tokens_in":15326,"tokens_out":12608,"duration_ms":101842,"concrete_test":"Re-run the fit using CMB distance priors computed self-consistently within the f(R,T) model: for each ε, evaluate R=sqrt(Ω_m)H0 DM(z*) and ℓ_A=π DM(z*)/rs(z*) using the model's H(z), and build a Gaussian likelihood from the Planck-measured R, ℓ_A, ω_b and their covariance. Compare the resulting ε distribution with the wCDM-prior result. If the central value shifts by more than ~0.01 or the uncertainty changes significantly, the wCDM transfer is the source of bias; if not, the quoted result is robust.","verdict_should_be":"UNCHANGED","load_bearing_attack":"Section IV.A uses the CMB distance priors (R, ℓ_A, ω_b) of Chen et al. [6], which were calibrated under the wCDM dark-energy model. These compressed likelihoods are not model-independent: R and ℓ_A depend on the expansion history through DM(z*) and rs(z*) (Eqs. 31–33), and the f(R,T) model changes the matter density evolution for ε≠0 (Eq. 27). The authors apply the wCDM priors to an f(R,T) model without checking that they are unbiased for nonzero ε. This is load-bearing because the CMB term is one of the two new datasets (with DESI BAO) that most tighten the ε constraint relative to the previous SNe-only analysis; a biased CMB prior could shift the posterior and artificially inflate agreement with ε=0. No full CMB likelihood or validation is provided.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper studies f(R,T)=R+λT^ε gravity in a flat FLRW universe, now including radiation and a wider set of cosmological datasets. The authors fit five parameters (ε, ξ, ω_b, H_0, M) to CMB distance priors, BAO measurements (DESI DR2 plus a compilation), cosmic chronometers, and Pantheon+SH0ES supernovae. They report a marginalized 68% constraint ε = 0.010^{+0.013}_{-0.021}, concluding that the model is consistent with the ΛCDM value ε=0. The analysis is supported by a table of best-fit parameters at representative ε values, a marginalized probability plot, and publicly available code and data links.","tokens_in":15631,"tokens_out":9235,"duration_ms":74727,"significance":"If the result holds, it is a useful and nontrivial constraint: the previously almost unconstrained power-law f(R,T) model is shown to be tightly restricted to near-ΛCDM behavior by current data, resolving a degeneracy found in the authors' earlier SNe-only analyses. The paper is commendably transparent in making code and data available, and the inclusion of radiation and correlated data is a clear improvement over prior work. The main risk is that the CMB likelihood is compressed to distance priors calibrated under wCDM and then applied to a modified-gravity model without validation; this directly affects the quoted ε posterior. The paper is otherwise clearly written and the statistical framework is standard.","major_comments":[{"comment":"The CMB term uses the distance priors (R, ℓ_A, ω_b) of Chen et al. [6], which were calibrated assuming wCDM. For ε≠0 the f(R,T) model has a different expansion history, and the compressed R and ℓ_A may be biased when compared against a likelihood derived under a different physical model. This issue is load-bearing because the CMB term is one of the new datasets that tightens the ε constraint. The authors should validate the use of these priors, e.g., by reproducing a full CMB likelihood for representative ε values (including Planck) or by demonstrating that the compressed constraints are insensitive to the calibration model within the prior range. Without this, the quoted 68% interval is not fully trustworthy.","section":"§IV.A, Eqs. (31)–(33)"},{"comment":"The BAO section states that 'certain data were excluded' following the suggestion of [11], but it does not specify which points, how many, or the precise criterion. This makes the analysis non-reproducible and could affect the fit. The exact excluded data points, the full BAO dataset, and the covariance matrices used should be tabulated in an appendix or provided in the repository.","section":"§IV.B"}],"minor_comments":[{"comment":"The symbol λ is used both for the action coupling and for the dimensionless function in Eq. (21). This is confusing; please rename one of them (e.g., use a barred or lowercase variant).","section":"§III, Eqs. (20)–(23)"},{"comment":"The table reports χ²_min but not the number of data points or degrees of freedom. Please include reduced χ² or the total number of data points so the reader can assess goodness of fit.","section":"§V, Table I"},{"comment":"The DESI DR2 data are only cited to [7]; the specific 9 BAO points and their correlations should be listed or made available in a machine-readable file.","section":"§IV.B"},{"comment":"The figure shows ρ(ε) but does not mark the quoted 68% interval. Adding the interval boundaries or describing how the interval was extracted from the normalized curve would improve clarity.","section":"§V, Fig. 1"},{"comment":"Typo: 'baryon accoustic oscillations' should be 'baryon acoustic oscillations'.","section":"§VI"}],"recommendation":"major_revision","confidential_remarks":"The central qualitative conclusion—that ε is small and consistent with 0—may well survive the CMB-prior concern, but the quantitative interval and the statement that the data 'strongly constrain ε' rest on unvalidated compressed likelihoods. I would be willing to accept after the authors demonstrate that a full CMB likelihood gives the same posterior, or otherwise quantify the bias. The paper is otherwise competently executed."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Punchline: this is a competent, incremental multi-probe fit that concludes f(R,T)=R+λT^ε is consistent with ε=0. The new result is the combined posterior ε=0.010^{+0.013}_{-0.021}, from CMB + BAO + CC + Pantheon+SH0ES. That is genuinely new relative to their SNe-only work, and the conclusion is defensible in direction.\n\nWhat is good: the authors extend their own formalism to include radiation, handle correlated data (Pantheon+ covariance, DESI correlations, M prior from SH0ES), and provide code and data links. The derivation of the Hubble parameter with radiation looks internally consistent. The discussion of model behavior, including the ξ prior bound from the theory's breakdown, is careful. They do not oversell; the conclusion notes ε=0 is well within the error bars.\n\nWhere the soft spots are: the CMB term uses compressed distance priors (R, ℓ_A, ω_b) from Chen et al., which were calibrated in wCDM. That is stated in Sec. IV.A, but the transfer to f(R,T) is not validated. Since these quantities depend on the expansion history through D_M(z*) and r_s(z*), and f(R,T) changes the matter density evolution for ε≠0, this could bias the ε posterior. The CMB is one of the two datasets that most tighten the constraint, so this is not a minor footnote. I don't think it overturns the central null result, but it should be addressed with either a full CMB likelihood or a validation that the priors are unbiased in this model. Second: Sec. IV.B excludes \"certain data\" per [11] without specifying which points. That is a reproducibility issue. The text also has OCR/notation glitches that slow checking. Minor: for ε>0 the model has a future breakdown; they assume it happens after today, which is acceptable but worth remembering.\n\nBottom line: for people working on f(R,T) phenomenology or multi-probe modified-gravity constraints, this is a useful consistency check. It deserves a serious referee; I would send it with a request to justify the CMB prior transfer and document the BAO exclusions. I would not desk-reject it.","headline":"A solid null result for f(R,T), but the wCDM-calibrated CMB priors are applied without validation and the BAO exclusions are under-documented; refereeing should focus there.","tokens_in":15995,"tokens_out":2918,"would_cite":true,"duration_ms":23860,"reading_group":"maybe","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":[],"pacs":[],"model":"deepseek-v4-flash","headline":"Combining four cosmological datasets pins the f(R,T) gravity exponent to ε=0.010^{+0.013}_{-0.021}, consistent with the standard model value ε=0.","keywords":["f(R,T) gravity","modified gravity","cosmological parameters","baryon acoustic oscillations","cosmic chronometers","Type Ia supernovae","ΛCDM","dark energy"],"falsifier":"Run the same five-parameter fit with the full Planck 2018 likelihood (or a likelihood emulator) in place of the Chen-Huang-Wang distance priors; if the resulting ε posterior shifts by more than the quoted error bars, the near-zero result was an artifact of the prior transfer.","tokens_in":15258,"feed_emoji":"🔭","tokens_out":4894,"duration_ms":37025,"temperature":0.7,"pith_summary":"The paper asks whether a specific family of modified-gravity theories, f(R,T)=R+λT^ε, can be distinguished from the standard cosmological model using current measurements. Combining CMB distance priors, baryon acoustic oscillations, cosmic chronometers, and the Pantheon+SH0ES supernova sample, it finds the best-fit modification exponent is ε=0.010 with a 68% interval of +0.013/−0.021. Because ε=0 (which recovers ΛCDM) lies inside this range, the analysis concludes that the data show no preference for this modification of gravity. The result matters because it turns a previously flexible model—supernova-only fits allowed almost any negative ε—into a tightly constrained one that behaves effectively like a cosmological constant.","feed_headline":"Modified-gravity exponent pinned to 0.010 by four datasets","feed_subtitle":"CMB, BAO, chronometers, and supernovae put ε=0 inside the error bars: no deviation from ΛCDM needed.","key_machinery":"The machinery is the one-parameter action f(R,T)=R+λT^ε, where T is the trace of the stress-energy tensor. Varying the action gives nonconserved stress-energy; the paper shows that an effective current J'^μ∝(ρ_m²/T)u^μ replaces the ordinary matter current, leading to a modified conservation law and an altered Hubble expansion. The dimensionless coupling ξ≡λ/(κ^{2ε}H_0^{2(1−ε)}) is the practical fit parameter, and its upper bound ξ_lim (which exists for ε>0) is what suppresses the probability density at larger ε, effectively cutting off the posterior near ε≈0.03.","core_discovery":"For the f(R,T)=R+λT^ε gravity model in a flat FLRW universe with radiation, the paper constructs the expansion history and a full χ² likelihood from CMB, BAO, cosmic chronometers, and Type Ia supernova data, including correlations. Marginalizing over nuisance parameters (ξ, ω_b, H_0, and M), it obtains the relative probability distribution for ε, centered at ε=0.010 with 68% bounds of +0.013 and −0.021. The standard value ε=0 is comfortably inside this range. The authors interpret this as no preference for the modified term, with the tight constraint driven mainly by the Pantheon+SH0ES supernovae and DESI DR2 BAO data.","pith_inferences":["Because the CMB distance priors were calibrated under wCDM, a full Planck-likelihood analysis could shift the central ε; readers should treat the quoted interval as conditional on that transfer being unbiased.","The derived H0 values (roughly 68–69 km/s/Mpc) fall on the Planck side of the Hubble tension, implying this model does not resolve the tension and could sharpen it if local measurements are used in the same fit.","The ξ_lim cutoff acts as an informative prior; using a different prior (e.g., allowing the theory breakdown at finite past redshift) would change the shape of the high-ε tail and could widen the error bar.","If future DESI BAO data continue to prefer small ε, the f(R,T)=R+λT^ε family becomes a candidate for a cosmological-constant-like limit rather than a dynamical dark-energy alternative."],"forward_implications":["A joint analysis of four independent cosmological datasets confines ε to |ε|≲0.03 at 68% confidence, ruling out the large-negative-ε region that supernova-only fits previously permitted.","The model's expansion history at the best fit is effectively indistinguishable from ΛCDM, so existing cosmological observations do not demand an f(R,T) modification.","If ε is truly zero, future BAO and supernova datasets should drive the ε posterior's central value toward zero and shrink its width.","The same formalism, including radiation, provides a reusable pipeline for testing other ε values or related modified-gravity forms against the same datasets."],"fun_headline_variants":["Four datasets peg modified gravity at ε=0.010, no ΛCDM shift","Gravity's f(R,T) exponent stays near zero: data says ε=0.010","CMB+BAO+chronometers+SNe: ε=0.010, consistent with standard model","Modified gravity exponent measured: ε=0.010, still no deviation"],"cache_read_input_tokens":2304,"weakest_assumption_plain":"The analysis relies on CMB distance priors that were calibrated under a different dark-energy model (wCDM) and applies them to f(R,T) expansion histories without a full CMB likelihood; if those priors are biased for ε≠0, the reported ε posterior would be biased.","fun_headline_variants_meta":{"raw":{"variants":["Four datasets peg modified gravity at ε=0.010, no ΛCDM shift","Gravity's f(R,T) exponent stays near zero: data says ε=0.010","CMB+BAO+chronometers+SNe: ε=0.010, consistent with standard model","Modified gravity exponent measured: ε=0.010, still no deviation"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000148,"raw_usage":{"total_tokens":970,"prompt_tokens":632,"completion_tokens":338,"prompt_tokens_details":{"cached_tokens":256},"prompt_cache_hit_tokens":256,"prompt_cache_miss_tokens":376,"completion_tokens_details":{"reasoning_tokens":255}},"tokens_in":376,"tokens_out":338,"duration_ms":3858,"temperature":1.0,"reasoning_tokens":255,"cache_read_input_tokens":256,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-01T21:10:48.840346+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Run the same five-parameter fit with the full Planck 2018 likelihood (or a likelihood emulator) in place of the Chen-Huang-Wang distance priors; if the resulting ε posterior shifts by more than the quoted error bars, the near-zero result was an artifact of the prior transfer.","supporting_citations":[],"review_version":1}