{"id":"d2fdd823-cf08-4ea5-9464-3e5a0373d43e","arxiv_id":"2412.14645","paper_version":2,"verdict":"CONDITIONAL","confidence":"HIGH","novelty_score":7.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":4,"one_line_summary":"Power-law correlations between fractional-maximum-mass radii and central EOS quantities provide a new, fast inversion of neutron star mass-radius curves into the dense-matter equation of state.","lead":"This paper develops a fast analytic method to recover the equation of state of dense matter from a neutron star's mass-radius curve. It reports sub-percent accuracy using power-law fits calibrated on hundreds of published equations of state.","discovery_kind":"new_method","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The paper's own Tables 2 and 3 contradict the headline 'typically 0.5% for all quantities': P, cs, and n RMS errors reach 1.3–5.7% at several fractional masses, even in-sample.","rationale":"Reading the paper in good faith, the intended contribution is a fast analytic inversion of a complete M-R curve into the dense-matter EOS, with semi-universal power-law fits accurate to 0.5%. For this to be true, the RMS errors reported in Tables 2 and 3 for E, P, cs, mu, and n at all fractional masses f must be near 0.5%. They are not. The pressure error at f=1 is 1.26%, sound speed at f=1 is 5.74%, and baryon density errors are ~1% across the board. These numbers come from the same 316-EOS training set, so the discrepancy is not about extrapolation to unknown EOSs; it is an internal inconsistency between the paper's headline and its own results. This is the single most load-bearing concern because the entire pitch of the paper rests on the 0.5% accuracy figure. If the figure is wrong, the claimed 'order-of-magnitude increase in precision' and the suitability of the method as an alternative to Bayesian inference are both undermined. The reader's weakest_assumption (representativeness of the training set) is related but distinct; even a perfectly representative training set would not rescue the claim, because the tables already violate it. The published Zenodo code and the Sun et al. (2024a) table make the check straightforward and reproducible. Since the paper could be revised by correcting the accuracy claims and adding out-of-sample tests, the reader's CONDITIONAL verdict remains appropriate rather than a full reject. The verdict should therefore be unchanged, but the required conditions should include correcting the abstract to report errors by quantity and mass point, and validating on an independent EOS family.","tokens_in":22071,"tokens_out":9837,"duration_ms":64328,"concrete_test":"Use the published Zenodo code (Sun et al. 2024b) and the Sun et al. (2024a) table of 316 EOSs to recompute the RMS errors (Eq. 17) for Table 2 (E, P, cs, P/E) and Table 3 (n, mu). In particular, verify <δP,f=1>, <δcs,f=1>, and <δn,f=1>. If the values reproduce the table entries (1.26%, 5.74%, 0.87%), the abstract's 'typically 0.5% for all quantities at all mass points' is contradicted by the paper's own data, and the accuracy claims must be revised to be quantity- and mass-dependent. If the values do not reproduce, the tables or code contain a computational error, which equally invalidates the central claim.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central quantitative claim (abstract, §5) is that the two-radius power-law fits (Eq. 22) achieve RMS accuracies 'typically 0.5% for all quantities at all mass points.' This is not supported by the paper's own tables, even before any out-of-sample or representativeness issue is considered. In Table 2, the RMS errors for pressure are 1.26% at f=1, 2.51% at f=3/5, and 3.13% at f=1/3; for sound speed, 5.74% at f=1. In Table 3, baryon density errors are 0.8–1.4% at essentially all f. These exceed the claimed 0.5% by factors of 2–10. The paper defines accuracy as RMS error (Eq. 17), so the discrepancy is unambiguous. Consequently, the abstract and §5 overstate the method's fidelity. The reader's concern about training-set representativeness is valid, but the more immediate problem is internal: even on the exact 316 EOSs used to calibrate the coefficients, the stated accuracy does not hold for several quantities and mass points. The method may still be useful at the few-percent level, but the headline claim as written is false.","agreement_with_reader":"partial"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper develops an analytic method to invert neutron-star mass-radius (M-R) curves into the underlying equation of state (EOS). The method uses power-law correlations between central values of energy density, pressure, sound speed, chemical potential, and baryon density at fixed fractions f of the maximum mass and two radii selected from the M-R curve (Eq. 22). The coefficients are fit to 316 hadronic EOSs from the Sun et al. (2024a) tabulation with Mmax >= 2 Msun. The authors claim that the two-radius fits achieve RMS accuracies of typically 0.5% for all quantities at all mass points, that the method works reasonably well for hybrid stars with first-order phase transitions, and that it provides an analytic alternative to Bayesian inversion. They also present fixed-mass-grid fits for application to observational data (Eqs. 29-30) and compare their inferred EOS with published Bayesian results for two NICER sources.","tokens_in":22336,"tokens_out":7612,"duration_ms":80081,"significance":"If the central accuracy claim and the semi-universality of the correlations were established, this would be a valuable fast analytic inversion tool, complementary to Bayesian methods. The paper is strengthened by the use of a large EOS database, an extensive comparison with the TPE approach and the fits of Ofengeim et al. (2023), the release of code in a Zenodo repository, and the tabulation of all fit parameters. However, the main quantitative claim is contradicted by the paper's own tables, the reported errors are purely in-sample with no cross-validation, and the out-of-sample evidence is limited to a single hybrid EOS. These issues currently undermine the paper's significance.","major_comments":[{"comment":"The central accuracy claim, 'typically 0.5% for all quantities at all mass points,' is contradicted by the paper's own tables. The RMS errors defined in Eq. (17) for the two-radius fits of Eq. (22) include <delta P> = 1.26% at f = 1, 2.51% at f = 3/5, and 3.13% at f = 1/3 (Table 2); <delta cs> = 5.74% at f = 1 (Table 2); and <delta n> approximately 0.8-1.4% at all f (Table 3). The statement in §3 that 'the average accuracy for P is better than 0.6% except for 1/3 <= f <= 3/5' is also inconsistent with the f = 1 entry of <delta P> = 1.26%. The abstract and §5 overstate the method's fidelity even on the training set.","section":"Abstract, §3, Tables 2-3"},{"comment":"The reported accuracies are in-sample RMS errors: the coefficients in Eq. (22) are fit by minimizing the chi-square of Eq. (23) over the same 316-EOS set, and the optimized pairs (gmin, hmin) are selected by the same minimization on the same data. No cross-validation, holdout set, or independent hadronic EOS compilation is used. The paper itself states in §5 that 'a future project will be to utilize several versions of parameterized EOSs to validate the fitting parameters found here.' Without such validation, the claim that the correlations are 'semi-universal' is not established; the (g,h) selection in particular could inflate apparent accuracy through overfitting.","section":"§3, Eqs. (22)-(23)"},{"comment":"The only out-of-sample test with first-order phase transitions is a single construction (BSk22 + MIT bag with B = 80 MeV fm^-3), discussed qualitatively as accurate to a few percent. Given that the abstract generalizes to 'hybrid stars with first-order phase transitions' and that this case is the sole evidence for robustness beyond the hadronic training set, the generalization is not yet supported. A systematic scan over bag constants and hadronic EOSs, or at least over a few representative transition pressures and densities, is needed to justify the abstract's claim.","section":"§5, Figs. 7-8"}],"minor_comments":[{"comment":"The value of aE for Mj = 1.3 Msun is listed as 0.08501, which is likely a typo for 0.8501; also, the values of bnu for Mj = 1.5 and 1.4 Msun are identical (-0.05371), which seems suspicious and should be checked.","section":"Table 4"},{"comment":"The phrase 'increases the RMF accuracies' appears to be a typo; it should be 'RMS accuracies.'","section":"§3"},{"comment":"The caption of Fig. 2 references Eq. (23) for the reconstructed points, but the fits are defined by Eq. (22), while Eq. (23) defines the chi-square; similarly, the caption of Fig. 5 references Eq. (23) where Eq. (29) is meant.","section":"Captions of Figs. 2 and 5"},{"comment":"The notation for the nuclear saturation energy density is inconsistent: the text uses both Es and E0; additionally, the hadronic EOS 'BSk22' is sometimes written as 'BSK22' in the captions of Figs. 7 and 8. Please unify the notation.","section":"§4 and §5"}],"recommendation":"major_revision","confidential_remarks":"The paper's central quantitative claim is internally inconsistent with its own tables, and the lack of out-of-sample validation for the hadronic case makes the 'semi-universal' assertion premature. I recommend major revision with a corrected accuracy statement and additional validation; if these are addressed, the paper could be suitable for publication."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"The genuinely new thing here is the two-radius power-law correlation at fractional maximum masses: fitting central E, P, cs, mu, n against Mmax and two optimized radii gets in-sample RMS errors down to ~0.2–0.5% for several quantities, an order of magnitude better than the single-radius or TPE fits used before. That is a real empirical result, and the authors are honest about the parts they couldn't reproduce in prior work. The tables are published, the Zenodo code is there, and the method is straightforward to implement. Credit where due: this is a useful tool for fast, parameterization-light EOS inference, and the comparison against Bayesian pipelines in Fig. 6 is a nice illustration of prior-driven systematics.\n\nThe main problem is the abstract's claim: \"typically 0.5% for all quantities at all mass points.\" The paper's own tables say otherwise. In Table 2, pressure RMS errors are 1.26% at f=1, 2.51% at f=3/5, 3.13% at f=1/3; sound-speed error is 5.74% at f=1; baryon density errors in Table 3 are 0.8–1.4% essentially everywhere. So even on the exact training set, the stated accuracy fails for several quantities by factors of 2–10. The stress-test note is right, and the discrepancy is not a matter of interpretation—the paper defines accuracy as RMS error in Eq. (17).\n\nThe second soft spot is the validation gap. Coefficients and the (g,h) radius choices are selected on the same 316 EOSs used to report the errors. No cross-validation is performed. The only out-of-sample test is one hybrid star with a first-order phase transition, where errors are a few percent. The paper explicitly defers proper validation to a \"future project\" in §5. That is a load-bearing limitation for the headline claim, though not for the method's existence. Also note the \"arbitrary M-R curve\" inversion requires Mmax, which is not directly observable; the fixed-mass alternative in §4 has ~5–9% errors.\n\nOverall, this is a solid, reproducible contribution that deserves a serious referee. The authors should be asked to correct the abstract, present per-quantity RMS errors without spin, and add at least one genuine out-of-sample test (e.g., leave-one-out cross-validation or a parameterized EOS family not in the training set). With that, the method would be a useful complement to Bayesian inference, not a replacement for it yet.","headline":"A genuinely new empirical inversion tool with in-sample sub-percent fits, but the abstract's \"0.5% for all quantities\" is contradicted by the paper's own tables, and no real out-of-sample validation backs it up.","tokens_in":22984,"tokens_out":2476,"would_cite":false,"duration_ms":17478,"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":"For ordinary hadronic equations of state, a neutron star's mass-radius curve can be analytically inverted to the dense-matter EOS with about 0.5% accuracy using power-law fits at two fractional maximum-mass radii.","keywords":["neutron stars","mass-radius relation","equation of state","TOV equations","analytic inversion","power-law correlations","hybrid stars","Bayesian statistics"],"falsifier":"Take an equation of state not in the training set, with a different crust treatment or a first-order phase transition, compute its M-R curve, apply Eq. (22), and compare the reconstructed P(E) with the true EOS; if the RMS deviation in pressure at any fractional-mass point exceeds the quoted 0.5-1% (or a few percent near a transition), the claimed semi-universality does not hold for that class.","tokens_in":21727,"feed_emoji":"⭐","tokens_out":8937,"duration_ms":90721,"temperature":0.7,"pith_summary":"The paper claims that the neutron star mass-radius (M-R) curve contains enough information to reconstruct the dense-matter equation of state (EOS) analytically, without Bayesian model fitting. Its central result is a set of power-law formulae that predict the central energy density, pressure, sound speed, chemical potential, and baryon number density of a star whose mass is a fixed fraction of the maximum mass, using only the maximum mass and two radii at fractional masses. For ordinary hadronic EOSs these predictions are accurate to roughly 0.5% root-mean-square at all mass points. If true, this gives a direct, prior-free alternative to Bayesian inference of the EOS from mass and radius measurements, turning an observed M-R curve into the pressure-energy-density relation.","feed_headline":"Mass-radius curves invert to the dense-matter EOS at 0.5% accuracy","feed_subtitle":"With just the maximum mass and two fractional-mass radii, one can reconstruct the equation of state to half a percent.","key_machinery":"The central object is the two-radius power-law fitting formula of Eq. (22): $G_f = a (M_{\\max}/M_\\odot)^{b} (R_g/10\\,\\mathrm{km})^{c} (R_h/10\\,\\mathrm{km})^{d}$, where $G_f$ denotes the central energy density, pressure, sound speed, chemical potential, or baryon density at a star of mass $f M_{\\max}$. The coefficients are determined by least-squares fitting on the 316-equation-of-state sample, and the two radii are chosen from an 11-point grid of fractional maximum masses to minimize the root-mean-square error. These formulae translate a point on the mass-radius curve into the thermodynamic state at the star's center, and interpolating over the 11 fractional masses reconstructs the whole equation of state up to the central values at the maximum mass.","core_discovery":"The authors establish that for a suite of 316 hadronic equations of state with maximum masses above two solar masses, the central energy density, pressure, sound speed, baryon chemical potential, and baryon number density of stars at masses equal to f times the maximum mass are determined by a power law in the maximum mass and in the radii at two chosen fractional maximum masses. Optimizing the choice of the two radii brings the fits to better than 1% accuracy and typically about 0.5% root-mean-square. Inverting an entire M-R curve point by point therefore yields the full pressure-energy-density relation, the central sound speed, and the chemical potential relation, with errors of order 1% or less. The same formulae applied to a hybrid star with a first-order phase transition reproduce the EOS away from the transition to a few percent and give the midpoint of the transition correctly, even though no hybrid EOS was used to train the fits. This amounts to an analytic, EOS-insensitive inversion of the Tolman-Oppenheimer-Volkoff equations.","pith_inferences":["If the semi-universality extends to equations of state beyond the training set, the same power-law coefficients could be applied to invert M-R curves of hybrid or quark-matter stars, but the paper's single hybrid test suggests accuracy degrades near the transition; a systematic study over many hybrid EOSs would quantify this failure mode.","The order-of-magnitude improvement gained by adding a second radius suggests that other pairs of nearly independent observables, such as radius plus tidal deformability or radius plus moment of inertia, may yield similarly sharp reconstructions of the central thermodynamic state.","If the inversion is truly prior-free, it could serve as a consistency diagnostic for Bayesian analyses: large discrepancies between inverted and Bayesian EOS bands would indicate that the Bayesian prior choice, not the data, is driving the result.","The paper notes a fundamental limit to single-point accuracy from the non-uniqueness of central density and pressure for a given (M,R) point, so the achievable precision of any inversion is bounded by how much structure the M-R curve carries, not just by the fitting function."],"forward_implications":["If the 0.5% correlations hold for real dense-matter equations of state, then measurements of several neutron star masses and radii can be converted directly into central pressure and energy-density estimates for each observed star.","The method provides a Bayesian-prior-free cross-check: EOS bands produced by parametric Bayesian analyses can be compared with the direct analytic inversion of the same M-R data, exposing prior-induced systematic differences.","The reconstruction also yields the central sound speed, baryon density, and chemical potential at each fractional mass, providing additional thermodynamic information that can be confronted with nuclear-theory predictions.","For hybrid stars with first-order phase transitions, the inversion smooths over the transition but still recovers the transition midpoint, so it can flag the presence of a strong phase transition when the reconstructed EOS shows an unusual softening or a density discontinuity.","The same approach is argued to extend to moments of inertia and tidal deformabilities, which are tightly correlated with mass and radius, broadening the set of observables that can be inverted analytically."],"supporting_citations":[{"why":"Derives the relativistic stellar structure equations whose mass-radius curve is the object of inversion.","marker":"Tolman 1934"},{"why":"Applies the Tolman equation to neutron matter, establishing the TOV relation that maps EOS to the M-R curve.","marker":"Oppenheimer & Volkoff 1939"},{"why":"Supplies the 316-equation-of-state tabulation used to fit the power-law coefficients and test the fits.","marker":"Sun et al. 2024a"},{"why":"Provides the SLy4 crustal EOS used as the common low-density equation of state for all models.","marker":"Chabanat et al. 1998"},{"why":"Gives the piecewise-polytrope fit of the SLy4 crust below 0.04 fm^-3 used in the TOV integrations.","marker":"Zhao & Lattimer 2022"},{"why":"Introduces the truncated perturbative expansion that motivates power-law correlations among maximum-mass quantities.","marker":"Cai et al. 2023a"},{"why":"Provides prior power-law-style fits that the paper compares against and improves with the two-radius approach.","marker":"Ofengeim et al. 2023"},{"why":"Earlier surface-to-center inversion scheme whose instabilities the analytic method seeks to avoid.","marker":"Lindblom 1992"}],"fun_headline_variants":["Analytic inversion turns M-R curves into EOS at 0.5%","From mass-radius to equation of state: 0.5% accuracy","Invert TOV: M-R curve yields dense-matter EOS precisely","Reconstruct neutron star EOS from M-R data with ~0.5% error","Power-law bridges neutron star radii and dense-matter EOS"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The fitted power-law correlations are assumed to be semi-universal beyond the 316 hadronic equations of state used to determine them, particularly for equations of state with first-order phase transitions, for which the paper tests only one hybrid model.","fun_headline_variants_meta":{"raw":{"variants":["Analytic inversion turns M-R curves into EOS at 0.5%","From mass-radius to equation of state: 0.5% accuracy","Invert TOV: M-R curve yields dense-matter EOS precisely","Reconstruct neutron star EOS from M-R data with ~0.5% error","Power-law bridges neutron star radii and dense-matter EOS"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.00088,"raw_usage":{"total_tokens":3838,"prompt_tokens":1015,"completion_tokens":2823,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":631,"completion_tokens_details":{"reasoning_tokens":2723}},"tokens_in":631,"tokens_out":2823,"duration_ms":16292,"temperature":1.0,"reasoning_tokens":2723,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-11T12:02:38.481065+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Take an equation of state not in the training set, with a different crust treatment or a first-order phase transition, compute its M-R curve, apply Eq. (22), and compare the reconstructed P(E) with the true EOS; if the RMS deviation in pressure at any fractional-mass point exceeds the quoted 0.5-1% (or a few percent near a transition), the claimed semi-universality does not hold for that class.","supporting_citations":[{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Derives the relativistic stellar structure equations whose mass-radius curve is the object of inversion."},{"cited_title":"R., & Volkoff, G","cited_arxiv_id":null,"evidence_quote":"Applies the Tolman equation to neutron matter, establishing the TOV relation that maps EOS to the M-R curve."},{"cited_title":"1992, Astrophys","cited_arxiv_id":null,"evidence_quote":"Earlier surface-to-center inversion scheme whose instabilities the analytic method seeks to avoid."}],"review_version":1}