{"id":"c0443be8-7f1b-4870-bf60-7a5f577a59e3","arxiv_id":"2412.03652","paper_version":2,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":5.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":3,"one_line_summary":"Axion energy loss can erase the blue loop stage in MESA models of 5 to 9 solar mass stars, giving new bounds on the axion-photon coupling from Cepheids and stronger projected bounds for heavier Cepheids.","lead":"This paper uses stellar evolution simulations to show that axion-like particles can erase the blue loop phase of intermediate-mass stars, and uses the observed Cepheid S Mus to place new bounds on the axion-photon coupling. The bounds are weaker than existing globular cluster limits, but the study identifies massive Cepheids as a promising future probe of axions.","discovery_kind":"extension","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Unexplained discrepancy in axion energy-loss G-table (Table 3.2 vs. Raffelt & Dearborn Table II) leaves all quantitative g10 thresholds unvalidated; even perfect mesh resolution cannot fix a miscalibrated epsilon_a.","rationale":"The reader's weakest assumption focuses on MESA interpolation as the source of finely tuned thresholds. That is a real concern, and the authors candidly flag it. However, the more load-bearing uncertainty lies one step earlier: the axion emission rate is the physical input that sets every threshold, and the paper itself notes that its G-table disagrees with the standard reference while using the same expressions. Without an independent check of that table, the quantitative bounds cannot be trusted even if the MESA mesh is made arbitrarily fine. This is not an ad hominem objection; it is an unvalidated numerical ingredient in the central calculation. The qualitative conclusion -- that increasing axion energy loss weakens or eliminates blue loops, and that heavier Cepheids are a promising future probe -- is plausible and supported by the monotonic behavior of the tracks, the mirror-principle discussion, and the hydrogen-profile analysis. Therefore the existing CONDITIONAL verdict remains appropriate rather than a rejection. The concrete test is straightforward and would resolve whether the discrepancy is a harmless normalization difference or a genuine error. If the table is correct, the quantitative results regain their claimed precision; if not, the reported constraints need to be recomputed. The absence of released code or data makes such an independent check essential.","tokens_in":28129,"tokens_out":10826,"duration_ms":107368,"concrete_test":"Independently compute G(y0,y1) from Eqs. 3.6-3.22 at selected grid points, for example (log y0, log y1) = (-1.0, 0.0), (0.0, 1.0), and (1.0, 1.5), using two independent numerical quadrature implementations, and compare against both Table 3.2 and Table II of [41]. If the authors' table differs from the independent evaluation by more than 5%, re-run the 6 Msun and 9 Msun blue-loop threshold searches with the corrected table; if the resulting S Mus bound moves outside its reported precision, the headline constraints are not robust. Also report the sign and magnitude of the difference relative to [41].","verdict_should_be":"UNCHANGED","load_bearing_attack":"The quantitative claim that blue loops vanish at specific g10 thresholds depends entirely on the Primakoff energy-loss rate epsilon_a(g10) inserted into MESA. In Section 3.2 the authors present Table 3.2 for G(y0,y1) and state that these values \"differ from those in Table II of [41], which use the same expressions,\" without reconciling the discrepancy or providing an uncertainty. Because epsilon_a is proportional to g10^2 times G in Eqs. 3.5, 3.16, 3.20, and 3.22, an unquantified factor of 1.2 in G shifts every threshold coupling by about sqrt(1.2) ~ 1.1, and a factor of 2 shifts thresholds by about 1.4. The S Mus constraint (g10 ~ 1.6-1.7), the 9 Msun bracket (0.83-0.835), and the linear fit in Eq. 5.1 all inherit this calibration risk. This is upstream of the MESA interpolation concern the authors themselves raise in Section 5.1: even with arbitrarily fine spatial resolution, the mapping from g10 to energy loss is only as reliable as the G-table. Since no MESA inlists, data, or code are released, the table cannot be audited from the paper alone.","agreement_with_reader":"partial"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper studies whether axion-photon energy loss can eliminate the blue-loop phase in intermediate-mass stars, using MESA r23.05.1 models with the Primakoff production rates of Raffelt and Dearborn. Using the core-potential criterion of Lauterborn, Refsdal, and Weigert, the authors find mass-dependent threshold couplings: a 6 Msun model loses its blue loop near g10 = 2 (with g10 about 1.757 for alpha_ov = 0.1), an 8 Msun model near g10 = 1.2253 versus 1.2254, and a 9 Msun model between g10 = 0.83 and 0.835. They translate these thresholds into constraints using the galactic Cepheid S Mus, whose dynamical mass is 6 Msun, reporting g10 about 1.6-1.7 for the mass and overshoot combinations in Table 4. A linear fit in Fig. 11 and Eq. (5.1) is used to project a tentative 12 Msun constraint in Fig. 18. The paper also analyzes the mirror principle, the hydrogen profile, and the sensitivity of the thresholds to convective overshoot, and argues that massive Cepheids are an important frontier for axion searches.","tokens_in":28390,"tokens_out":7995,"duration_ms":81789,"significance":"If the quantitative thresholds are correct, the paper would introduce a genuinely new stellar-evolution probe of the axion-photon coupling, complementary to globular-cluster R and R2 constraints and potentially much stronger for massive Cepheids. The authors deserve credit for implementing the full non-degenerate plus degenerate Primakoff formalism, for verifying loop suppression directly on HR tracks (Fig. 10), and for being unusually candid about the limitations of their simulations, including the possible MESA interpolation artifact in Sec. 5.1 and the lack of theoretical support for the linear fit. However, the claimed constraints currently lack uncertainty quantification, the G-table calibration is unresolved, and the S Mus bound is not a single well-defined number. In its present form the paper is a compelling proof of principle rather than a calibrated constraint.","major_comments":[{"comment":"Section 3.2 and Table 1: the paper states that its computed values of G(y0,y1) differ from Table II of Raffelt and Dearborn [41] even though the same expressions are used, but it does not reconcile the discrepancy or assign an uncertainty. Since the energy-loss rates in Eqs. (3.5), (3.16), (3.20), and (3.22) are all proportional to g10^2 times this G function, an unquantified multiplicative error in G translates directly into a shift of every threshold coupling by the square root of that factor. The 9 Msun bracket, the S Mus values in Table 4, and the fit in Eq. (5.1) all inherit this calibration risk. Please provide a comparison with the original table, explain the source of the differences, or quantify the resulting uncertainty in g10.","section":"3.2 (Table 1)"},{"comment":"Section 5.1, Fig. 9, and the surrounding text: the reported thresholds resolve differences as small as Delta g10 = 10^-4 (for example, g10 = 1.2253 versus 1.2254 for the 8 Msun model), while the authors themselves write that the elimination 'may be an artifact of the stellar evolution interpolation in MESA, rather than an accurate representation of the effect of axions.' The quantitative constraints are anchored on this sharpness. Please supply convergence tests in mesh resolution, timestep control, the interpolation scheme for the G table, and the treatment of y0 values outside the tabulated range, and state a conservative coarse threshold with an error bar. Without such support, the fine-grained values in Fig. 11, Eq. (5.1), and Table 4 cannot be taken as reliable.","section":"5.1"},{"comment":"Section 5.2, Tables 3 and 4: the S Mus constraint is obtained from three mass plus overshoot combinations (6.0/0.2, 6.4/0.1, and 6.6/0.0) that all reproduce the nominal luminosity log L/Lsun = 3.54, but only the first is compatible with the dynamical mass of S Mus quoted in Table 2. The resulting g10 values span a factor of about five, from 0.35 to 1.7, so the paper does not actually deliver a single conservative bound. The authors acknowledge that a proper scan over {M, alpha_ov, g10} is required; please perform that scan or explicitly present the 6.4 and 6.6 Msun rows as illustrative model-dependence studies rather than as constraints from S Mus.","section":"5.2 (Tables 3 and 4)"},{"comment":"Eq. (5.1) and Fig. 18: the linear fit is based only on masses up to 9 Msun and is extrapolated to 12 Msun. The authors are unable to simulate a blue loop at 12 Msun even without axions, and they admit they 'cannot justify the linear fit in Eq. 5.1 at a theoretical level.' The orange contour in Fig. 18 is therefore a speculative projection, not a constraint. Please move this material to a clearly labeled prospects section and remove any impression in the abstract or conclusions that a 12 Msun bound has been derived.","section":"5.1 and Appendix A"},{"comment":"Throughout Sections 4.4-5.3, threshold couplings are quoted to three or four significant digits without any uncertainty from input physics (metallicity, alpha_MLT, the nuclear network, semiconvection) or from numerical resolution. Given the paper's own characterisation of the blue loop as a 'magnifying glass', a quantitative statement of which systematic variations move the thresholds by more than the quoted digits is necessary before any of the numbers in Tables 4 or Eq. (5.1) can be used as bounds. At minimum, the authors should state the grid resolution and timestep controls used for the fiducial runs and show that the thresholds are stable under reasonable variations.","section":"4.4-5.3"}],"minor_comments":[{"comment":"In the row labelled log y1 = 0.0 and the column log y0 = 1.0, the entry reads '0.0.039'; this appears to be a typo for 0.039.","section":"Table 1"},{"comment":"The table is referred to as 'Table 3.2' in the text but is captioned 'Table 1'; the numbering should be made consistent.","section":"3.2"},{"comment":"The formula h = e^{const.·Delta X} Delta m is unclear as written and appears inconsistent with the statement that Delta m -> 0 implies h -> 1; please clarify the intended functional form.","section":"4.1, Eq. (4.1)"},{"comment":"The 0.07% consistency check quoted for the extended G table concerns final stellar ages only; it does not verify that the interpolated table reproduces the blue-loop thresholds, which is the quantity actually used in the paper.","section":"3.2"},{"comment":"The phrase 'The determination of masses by dynamical orbital methods that are unaffected by axion physics fixes the trajectory of these candidates on the HR diagram' is grammatically incomplete; it should say that the mass determination, combined with the HR-diagram position, fixes the trajectory.","section":"2"},{"comment":"In the conclusions, 'strong constrains' should read 'strong constraints'.","section":"6"},{"comment":"The vertical axis label 'log Phi_c' should use the same symbol phi_c as in Eq. (1.1) for consistency.","section":"Figures 6 and 14-15"}],"recommendation":"major_revision","confidential_remarks":"I am sympathetic to the qualitative message, and the paper does a genuine service by directing attention to massive Cepheids as axion probes. My recommendation of major revision is driven by the G-table discrepancy, the unresolved dependence of the thresholds on MESA interpolation, and the fact that the S Mus bound is not a single robust number. These issues are fixable in a revision: reconcile or propagate the G-table uncertainty, add convergence tests, and reframe Tables 3-4 and the 12 Msun projection as illustrations rather than constraints. I would not recommend rejection because the underlying physical trend is clearly visible in the tracks and the authors are transparent about the limitations."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Dear colleague,\n\nThe one thing you should know: this is a careful, unusually candid numerical study, and the qualitative result—axion energy loss suppresses blue loops in MESA models, with heavier stars affected at smaller couplings—is credible and worth keeping in mind. The quantitative bounds, however, carry more uncertainty than the headline numbers suggest. I would not yet quote g10 values from this paper in a constraints plot.\n\nWhat is actually new: the authors implement the degenerate-medium Primakoff contribution in MESA (previous Cepheid work used only the non-degenerate piece), scan convective overshoot for 6 and 8 M_sun models, use the core potential as a diagnostic, and make S Mus, a Cepheid with a dynamical mass, their benchmark. They also cite and distinguish the earlier Friedland-Giannotti-Wise and Choplin et al. studies rather than burying them. The published caveats are honest: Section 5.1 openly says the fine elimination thresholds may be a MESA interpolation artifact; Appendix A says the 12 M_sun projection comes from neither simulation nor observation; Section 5.2 discusses the mass discrepancy problem. That transparency earns real credit.\n\nThe soft spots are real, though. The stress-test concern about the G-table is on target. Table 3.2 is supposed to implement the Raffelt-Dearborn expressions, but the paper states the values differ from Table II of [41] without reconciling the difference or giving an uncertainty. Since epsilon_a is proportional to g10^2 times G, an unquantified factor of 1.2–2 in G shifts every threshold by roughly the square root of that factor. This is upstream of the MESA interpolation issue: even a perfectly resolved mesh cannot fix a miscalibrated energy-loss rate. Second, the finely tuned values (8 M_sun: 1.2253 vs 1.2254; 9 M_sun: 0.83 vs 0.835) are presented with four significant digits despite the authors' own warning. They say it is not a precision study, and I agree—so the numbers should be displayed with that in mind, or rounded. Third, the linear fit Eq. 5.1 is used to project a 12 M_sun constraint, and the authors explicitly say they cannot justify it theoretically. That is fine as a teaser, but it should not be labeled a bound. Finally, no MESA inlists, data, or code are released, so the thresholds cannot be audited independently.\n\nWho is this for: people working on axion searches in stellar systems and on Cepheid/blue-loop modeling. The paper makes a credible case that massive Cepheids with dynamical masses are a promising frontier, even though the current S Mus constraint is weaker than existing globular-cluster limits. The central argument holds up qualitatively; the quantitative layer needs work.\n\nRecommendation: send to a serious referee. Require, at minimum, a reconciliation or uncertainty estimate for the G-table, released code/data or a detailed reproducibility appendix, rounded or caveated threshold values, and a clearer separation between simulated constraints and extrapolated projections.","headline":"Qualitatively credible and unusually candid, but the quantitative axion bounds are not yet reliable enough to quote; worth a serious referee with required fixes to the energy-loss table and reproducibility.","tokens_in":28979,"tokens_out":3584,"would_cite":true,"duration_ms":36644,"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":"This paper argues that axion energy loss can erase the blue loop of intermediate-mass stars, turning the observed existence of Cepheids like S Mus into a constraint on the axion-photon coupling.","keywords":["axions","axion-photon coupling","blue loops","Cepheids","core potential","Primakoff process","instability strip","convective overshoot"],"falsifier":"Run the same stellar-structure simulations with substantially finer spatial and temporal resolution, or with an independent stellar evolution code, and check whether the blue-loop suppression at $g_{10}=1.2253$ versus $1.2254$ for the $8\\,M_\\odot$ model and between $g_{10}=0.83$ and $0.835$ for the $9\\,M_\\odot$ model persists. If the threshold moves by more than the reported $\\Delta g_{10}\\sim 10^{-4}$ or disappears entirely, the fine-grained constraints are numerical artifacts even if the qualitative trend survives. Alternatively, determine a dynamical mass for a galactic Cepheid near $9\\,M_\\odot$: if such a star is found to pulsate with a coupling above the predicted threshold, the central claim is falsified.","tokens_in":27889,"feed_emoji":"⭐","tokens_out":7348,"duration_ms":69891,"temperature":0.7,"pith_summary":"This paper tries to show that axions, hypothetical light particles that couple to photons, act as an extra energy-loss channel in intermediate-mass stars, and that this loss can on its own suppress the blue loop, the hot excursion in the Hertzsprung-Russell diagram whose crossing defines classical Cepheids. If true, the very existence of a Cepheid with a dynamically measured mass becomes a limit on how strongly axions can couple to photons. The authors simulate models from 4 to 9 solar masses with and without axion emission and identify the smallest coupling at which the loop fails to reach the instability strip. Using the 6-solar-mass Cepheid S Mus as a benchmark, they report a constraint on the dimensionless coupling $g_{10}$ between about 1.6 and 1.7 depending on the overshoot prescription; heavier stars would give stronger constraints.","feed_headline":"Axion losses can erase a star's blue loop","feed_subtitle":"If a measured Cepheid like S Mus still pulses, the axion-photon coupling must stay below g10 ≈ 1.6–1.7.","key_machinery":"The load-bearing diagnostic is the core potential, $\\phi_c = h(\\Delta X,\\Delta m)\\,M_c/R_c$, where $M_c$ and $R_c$ are the core mass and radius and $h$ encodes how the hydrogen-abundance jump and shell width change across the hydrogen-burning front. The paper's criterion, inherited from earlier stellar-structure theory, is that a blue loop occurs only if $\\phi_c$ falls below a simulation-determined critical value $\\phi_{\\rm crit}$. Axion energy loss shrinks the core radius, keeping $\\phi_c$ high enough to prevent the loop. This single scalar metric converts a complex evolutionary track into an on/off switch, which is what allows the coupling threshold to be read off from whether the loop reaches the red edge of the instability strip; the underlying physical mechanism is the mirror principle, in which contraction of the core drives expansion of the envelope across the burning shell, and axions blunt it by limiting core growth.","core_discovery":"The central claim is that axion emission through the Primakoff process suppresses the growth of the helium core during the post-main-sequence phase, keeping the core potential $\\phi_c$ above the critical value $\\phi_{\\rm crit}$ needed for the core to expand and the envelope to contract along a blue loop. For each stellar mass there is a sharp threshold coupling: the loop of a $6\\,M_\\odot$ model disappears at $g_{10}=2$, the $9\\,M_\\odot$ loop disappears between $g_{10}=0.83$ and $0.835$, and for the $8\\,M_\\odot$ model the loop is present at $g_{10}=1.2253$ but gone at $1.2254$. Because heavier stars have hotter cores, they lose their loops at smaller couplings even though their no-axion loops are more robust. Requiring that a real observed Cepheid still cross the instability strip therefore bounds $g_{a\\gamma\\gamma}$; the benchmark constraint from S Mus is $g_{10}\\simeq 1.6$–$1.7$ for the mass and overshoot combinations that reproduce its observed luminosity. A linear fit, $g_{10}=(-0.28\\pm 0.02)\\,M/M_\\odot+(3.5\\pm 0.1)$, is used to project a much stronger constraint for a $12\\,M_\\odot$ Cepheid, although the paper itself marks that projection as tentative because no blue loop can currently be simulated at that mass.","pith_inferences":["Editorial extension: the authors' own caveat in Section 5.1 implies a direct numerical experiment—recompute the 8 and 9 solar-mass thresholds with finer timesteps or an independent stellar evolution code—and if the $g_{10}$ threshold shifts by more than the reported $10^{-4}$ level, the quantitative bounds should be quoted with one or two significant digits, not four.","Editorial extension: extrapolating the linear fit to a $12\\,M_\\odot$ Cepheid gives $g_{10}\\simeq 0.14$, which would approach or surpass the globular-cluster $R$ and $R_2$ constraints shown in the paper's Figure 13; a dynamically confirmed Cepheid near that mass would therefore be a high-value target for an independent axion limit.","Editorial extension: the non-monotonic overshoot dependence (for $8\\,M_\\odot$, $\\alpha_{\\rm ov}=0.2$ yields a stronger bound than $\\alpha_{\\rm ov}=0.1$) suggests that high overshoot acts in the same qualitative direction as axion emission, pushing the core potential toward its critical value; empirical overshoot determinations from binary eclipsing systems could therefore be fed directly into the "],"forward_implications":["A Cepheid with a dynamically measured mass becomes a one-sided limit on $g_{a\\gamma\\gamma}$ once its blue loop is required to cross the instability strip.","Heavier Cepheids give stronger limits: the simulated $9\\,M_\\odot$ threshold is $g_{10}\\simeq 0.83$–$0.835$, roughly twice as strong as the S Mus bound.","Axion emission shortens the helium-burning phase and moves the star onto the asymptotic giant branch earlier, so the phase duration itself carries a separate, population-level signature.","Axions do not move the Cepheid loop vertically in the HR diagram; they either allow or suppress the loop, leaving the Cepheid mass-discrepancy problem essentially unchanged."],"supporting_citations":[{"why":"Supplies the core-potential criterion $\\phi_c<\\phi_{\\rm crit}$ and the critical values used to diagnose whether a blue loop occurs.","marker":"[8]"},{"why":"Introduces the core-potential criterion relating core potential to the occurrence of blue loops; the paper's diagnostic is built on it.","marker":"[9]"},{"why":"Gives the dynamically determined mass of S Mus ($6\\,M_\\odot$) used as the benchmark Cepheid for the axion constraint.","marker":"[18]"},{"why":"Provides the corrected axion energy-loss normalization and a previous stellar-evolution study whose approach the paper contrasts with its own.","marker":"[25]"},{"why":"First simulation-based constraints on axions from Cepheids; the paper extends this work by including degenerate-medium emission and convective overshoot.","marker":"[26]"},{"why":"Provides the Primakoff axion production treatment with plasma-frequency and Debye screening effects that is implemented in the simulations.","marker":"[41]"},{"why":"Explains why blue loops are inhibited above roughly $9\\,M_\\odot$ (limited inward penetration of the convective envelope), setting the simulated mass ceiling.","marker":"[44]"},{"why":"Provides the observed luminosity of S Mus used to fix the benchmark mass-overshoot combinations in Table 4.","marker":"[52]"},{"why":"Supplies the quoted effective temperature of S Mus, locating it within the instability strip for the benchmark models.","marker":"[53]"}],"fun_headline_variants":["Axions can erase a star's blue loop","Blue loops vanish beyond a critical axion strength","Cepheid S Mus sets new axion-photon bounds","Axion emission kills Cepheid blue loops","Axion coupling capped by galactic Cepheid"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The load-bearing premise is that the sharp coupling thresholds at which the blue loop disappears are real stellar physics rather than numerical artifacts of the simulation's interpolation across the hydrogen-burning shell.","fun_headline_variants_meta":{"raw":{"variants":["Axions can erase a star's blue loop","Blue loops vanish beyond a critical axion strength","Cepheid S Mus sets new axion-photon bounds","Axion emission kills Cepheid blue loops","Axion coupling capped by galactic Cepheid"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.00035,"raw_usage":{"total_tokens":2033,"prompt_tokens":1188,"completion_tokens":845,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":804,"completion_tokens_details":{"reasoning_tokens":771}},"tokens_in":804,"tokens_out":845,"duration_ms":8463,"temperature":1.0,"reasoning_tokens":771,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-11T22:14:32.614749+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Run the same stellar-structure simulations with substantially finer spatial and temporal resolution, or with an independent stellar evolution code, and check whether the blue-loop suppression at $g_{10}=1.2253$ versus $1.2254$ for the $8\\,M_\\odot$ model and between $g_{10}=0.83$ and $0.835$ for the $9\\,M_\\odot$ model persists. If the threshold moves by more than the reported $\\Delta g_{10}\\sim 10^{-4}$ or disappears entirely, the fine-grained constraints are numerical artifacts even if the qualitative trend survives. Alternatively, determine a dynamical mass for a galactic Cepheid near $9\\,M_\\odot$: if such a star is found to pulsate with a coupling above the predicted threshold, the central claim is falsified.","supporting_citations":[{"cited_title":"Lauterborn, S","cited_arxiv_id":null,"evidence_quote":"Introduces the core-potential criterion relating core potential to the occurrence of blue loops; the paper's diagnostic is built on it."},{"cited_title":"Cepheid Masses: FUSE Observations of S Mus","cited_arxiv_id":"astro-ph/0607489","evidence_quote":"Gives the dynamically determined mass of S Mus ($6\\,M_\\odot$) used as the benchmark Cepheid for the axion constraint."},{"cited_title":"Effects of axions on Population III stars","cited_arxiv_id":"1707.01244","evidence_quote":"Provides the corrected axion energy-loss normalization and a previous stellar-evolution study whose approach the paper contrasts with its own."},{"cited_title":"Raffelt and D.S.P","cited_arxiv_id":null,"evidence_quote":"Provides the Primakoff axion production treatment with plasma-frequency and Debye screening effects that is implemented in the simulations."},{"cited_title":"Walmswell, C.A","cited_arxiv_id":null,"evidence_quote":"Explains why blue loops are inhibited above roughly $9\\,M_\\odot$ (limited inward penetration of the convective envelope), setting the simulated mass ceiling."},{"cited_title":"Binary Cepheids: Separations and Mass Ratios in $5\\,M_\\odot$ Binaries","cited_arxiv_id":"1307.7123","evidence_quote":"Provides the observed luminosity of S Mus used to fix the benchmark mass-overshoot combinations in Table 4."},{"cited_title":"Böhm-Vitense, N","cited_arxiv_id":null,"evidence_quote":"Supplies the quoted effective temperature of S Mus, locating it within the instability strip for the benchmark models."}],"review_version":1}