{"id":"4ae2cdc4-cec9-4ea5-81c3-9f39d9ccc949","arxiv_id":"2501.07119","paper_version":3,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":5.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":23,"one_line_summary":"Massive dark stars could reproduce the JWST star formation efficiency excess at z~11-14, but the relic black holes they leave behind would violate MACHO limits, so they cannot be the dominant explanation.","lead":"This paper tests whether dark stars, hypothetical stars powered by dark matter annihilation, can explain the surprisingly high star formation efficiency seen in JWST's most distant galaxies. It finds they could in principle, but the black holes left behind would be too numerous, so other explanations like Population III stars are preferred.","discovery_kind":"new_application","skeptic_critique":{"model":"deepseek-v4-flash","headline":"MACHO exclusion in Fig. 3 uses z~14 halo mass fractions without diluting for halo growth to z=0, overstating relic BH abundances by an order of magnitude or more.","rationale":"The paper's main novel claim is that dark stars with the top-heavy IMF inferred from the JWST UV LF fits are excluded by MACHO observations. I examined the mapping from the fitted dark-star SFE to the MACHO halo fraction. Eq. (11) transforms ϵDS,eff into ψ(m) = ϵDS,eff fb times a normalized IMF weight, i.e., the BH mass produced in the z~14 halo divided by that same halo's mass. However, the constraints used in Fig. 3 apply to present-day structures. The halo that hosts a z~14 dark-star-forming galaxy is much smaller than its z=0 descendant, so the same relic BH mass represents a smaller fraction of the present-day halo. The paper does not apply any correction for this mass growth. This is not a subtle normalization issue; the growth factor is typically an order of magnitude or more for the ⪆10^10 M_sun halos relevant to the JWST excess, and even for UFDs the progenitor mass is only a fraction of the present mass. Thus the solid exclusion boxes in Fig. 3 are systematically too high. If corrected, the exclusion likely weakens substantially. The reader's weakest assumption concerned the attribution of the residual UV luminosity to dark stars/Pop III and the fixed Pop II SFE; my concern is distinct and more direct to the paper's central claim, because even granting the attribution, the MACHO comparison as presented does not yield the claimed exclusion. I therefore recommend a condition requiring the authors to compute the present-day BH fraction via merger trees before the MACHO argument is accepted. The fitting and spectrum analysis may be sound, but the headline conclusion is not yet established.","tokens_in":26090,"tokens_out":15992,"duration_ms":164243,"concrete_test":"Build extended Press-Schechter merger trees from z=14 to z=0, and at each z=14 node add BH mass M_BH = ϵDS,eff fb M_h using the best-fit z=14 SFE parameters (with and without dust). At z=0, tally the BH mass fraction in bins of M_BH for the descendant halos (or globally over all halos). Re-plot Fig. 3 with these evolved fractions. If the evolved boxes fall below the Oguri/Brandt/Graham-Ramani constraints by more than a factor of ~10, the central MACHO exclusion claim is an artifact of ignoring halo growth.","verdict_should_be":"CONDITIONAL","load_bearing_attack":"Section 3's central exclusion argument relies on Eq. (11), ψ(m) = ϵDS,eff fb (ξ0 ϕ(m)m^2/Mtot), which gives the mass of relic BHs as a fraction of the z~14 host halo mass. The constraints in Fig. 3 (Oguri et al. 2018; Brandt 2016; Graham & Ramani 2024) instead bound the present-day fraction of compact objects in local halos or along lens lines of sight. Between z~14 and z=0, a host halo grows by accretion and mergers by a factor typically ≥10–1000, so the relic-BH fraction is diluted by M_h(z=14)/M_h(z=0). The paper never applies this dilution: the solid boxes in Fig. 3 plot ψ as if the present halo mass were the z~14 halo mass. For the JWST-bright halos at z~14 (M_h ≈ 10^10–10^11 M_sun), the z=0 descendants are about 10^12–10^14 M_sun, so the overestimate is one to three orders of magnitude. If the dilution is accounted for, the boxes can move below the MACHO constraints, removing the stated reason to exclude very massive dark stars as the dominant SFE source.","agreement_with_reader":"disagree"},"referee_report":{"model":"deepseek-v4-flash","summary":"Using the JWST/HST UV luminosity functions at z≈4–14, the authors fit a two-component star-formation-efficiency (SFE) model: a Pop II component calibrated at z=4–9 and a dark-star (or Pop III) power-law component that accounts for the residual at z=11–14. They report that dark stars with M≳10^3 M_sun can reproduce the z≈11–14 SFE excess, with WIMP masses from tens of GeV to a few TeV. They then convert the best-fit z≈14 dark-star SFE into a relic black-hole mass fraction and compare it with MACHO constraints, concluding that dark stars with a top-heavy IMF extending to 10^4–10^5 M_sun are excluded and that only a small fraction of the excess can come from very massive dark stars, leaving Pop III stars as a plausible alternative.","tokens_in":26623,"tokens_out":12421,"duration_ms":125288,"significance":"The paper assembles a broad compilation of UV LF data, uses Bayesian nested-sampling fits with evidence comparisons for dust/no-dust cases, and anchors the final constraint in external MACHO observations. These are real strengths: the central result, if robust, would connect the JWST SFE excess to dark-star physics and set limits on the dark-star IMF and WIMP parameter space. However, the quantitative exclusion is currently overstrong as presented. The MACHO comparison omits the dilution of the high-redshift black-hole fraction by subsequent halo growth, and it uses best-fit SFE values rather than the full posterior distributions. Because these issues bear directly on the paper's main conclusion, the contribution is more conditional than the abstract suggests.","major_comments":[{"comment":"The relic-BH fraction ψ(m) in Eq. (11) is normalized to the z≈14 host-halo mass through ε_DS,eff f_b, but the constraints plotted in Fig. 3 (Oguri et al. 2018; Brandt 2016; Graham & Ramani 2024) constrain present-day compact-object fractions in local halos or along lens lines of sight. No account is taken of the growth of the host halos between z≈14 and z=0. For the JWST-bright halos considered in the fit (M_h≈10^10–10^11 M_sun at z≈14), the z=0 descendants are typically 10^12–10^14 M_sun, so the solid boxes in Fig. 3 are inflated by roughly one to three orders of magnitude. Because this comparison is the basis for the statement that relic BHs are 'too abundant,' the exclusion is not established as written; the authors should either propagate each z≈14 halo to its z=0 descendant or compare a cosmic comoving BH density against the appropriate cosmological limits.","section":"Section 3, Eq. (11) and Fig. 3"},{"comment":"The MACHO exclusion uses the best-fit values of ε_DS,eff only. At z≈13–14 the posterior distributions are broad: with dust, ε_DS = 0.063^{+0.069}_{-0.045} (z=13) and 0.103^{+0.066}_{-0.060} (z=14); without dust, 0.066^{+0.058}_{-0.045} and 0.090^{+0.064}_{-0.052}, with best-fit values that sometimes differ from the posterior mode. The authors themselves note that the z≈13–14 UV LF samples contain very few points. The conclusion that a given IMF is 'excluded by MACHO constraints' should therefore be a posterior statement (for example, the fraction of posterior samples lying above the limits), not a binary statement based on a maximum-likelihood point.","section":"Section 2.3 and Table 2"},{"comment":"The dark-star component is fitted to the same z=11–14 UV LF data after fixing the Pop II SFE to its z=4–9 values, so the reproduction of the SFE excess is a fit rather than an independent prediction. The inferred ε_DS,eff and hence the MACHO bound depend on the assumptions that the Pop II SFE has no redshift evolution beyond z≈9 and that the entire residual is described by the fDS power law of Eq. (6). Redshift evolution of fS, dust attenuation, IMF variations, or AGN contamination in the residual would change ε_DS,eff and weaken or remove the exclusion. The paper should quantify how much fS evolution (or dust) is needed to eliminate the dark-star component, or present the dark-star reconstruction as an upper envelope rather than a best-fit scenario.","section":"Section 2.3, Eqs. (2) and (6)"},{"comment":"The conclusion that there is 'no appropriate mass range' depends on the lower bound M_DS ≳ 10^3 M_sun derived from the single-object spectral fit to JADES-GS-z13-0. This fit assumes a blackbody dark-star component, a fixed young stellar template, and an IGM damping-wing treatment, and the resulting M_DS–mχ posterior is not propagated into the MACHO comparison. Since the 500–945 M_sun window is the only one that survives the MACHO constraints, systematic uncertainties in the spectral fit could reopen that window; the paper should either quantify these systematics or soften the claim that dark stars are excluded in the full allowed mass range.","section":"Section 3, Fig. 10, and Appendix C"}],"minor_comments":[{"comment":"The sentence 'The solid boxes in Figure 3 illustrate the best-fitted fractions of Pop III stars' is inconsistent with Fig. 3, where the Pop III model is shown as the orange dashed box; please correct.","section":"Section 3, paragraph after Fig. 3"},{"comment":"Typo: 'dar star' should be 'dark star'.","section":"Section 1, penultimate paragraph"},{"comment":"Typo: 'interply' should be 'interplay'.","section":"Appendix C"},{"comment":"The WIMP mass entries use 'Gev' instead of 'GeV', and the labels 'DS, w Cap' and 'DS, wo Cap' are not explained in the table footnote; please define the with/without-capture distinction.","section":"Table 3"},{"comment":"The reference list contains two identical entries for Iocco et al. (2008), one under MNRAS and one under Mon. Not. Roy. Astron. Soc.; please merge them.","section":"References"},{"comment":"Please clarify whether the 'without capture' case yields no posterior at all or simply no overlapping region in Fig. 10, since the H-R diagram in Fig. 9 includes without-capture tracks.","section":"Appendix D"}],"recommendation":"major_revision","confidential_remarks":"The paper addresses a timely question and the data compilation is useful. My main concern is that the headline exclusion is presented with more certainty than the current analysis supports; the halo-growth dilution issue in particular can move the constraint by orders of magnitude, so the revision should be substantial. I do not see a problem with novelty or attribution, and the paper is within the journal's scope."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Here's my take on Lei et al. The paper asks whether dark stars can explain the JWST SFE excess at z~11-14, and claims that while a population of >10^3 Msun dark stars could reproduce the excess, the relic BHs from a top-heavy IMF would violate MACHO constraints, so only a small fraction of the excess can come from dark stars. The MACHO part is the key new result, and it has a load-bearing problem: the relic BH fraction is computed relative to the z~14 host halo and then compared to local constraints without accounting for halo growth to z=0. For the JWST-bright halos (Mh ~1e10-1e11), the descendants are ~1e12-1e14, so the BH fraction is diluted by 1-3 orders of magnitude. The stress-test note is right. Once you include the dilution, the exclusion boxes move down and could sit below the current MACHO limits, removing the stated reason to exclude very massive dark stars. This is a serious flaw in the central argument.\n\nWhat the paper does well: it carefully compiles the available UV LF data at 4≤z≤14, fits a two-component SFE model with a power-law dark star SFE, and shows that dark stars and Pop III stars have very similar KUV conversion factors, which is useful and not something I had seen so clearly. The spectrum fit to JADES-GS-z13-0 and the derived constraint that the dark star mass must exceed ~10^3 Msun is a nice extra.\n\nOther soft spots, in proportion: the 'reproduction' of the excess is a fit, not a prediction—the residual UV luminosity is assigned to dark stars by construction, with Pop II SFE fixed from z~4-9 and no redshift evolution. Also, the MACHO boxes use only best-fit SFE values, not the broad posteriors at z=13-14. The top-heavy IMF slope and complete collapse to BHs are assumptions, not results.\n\nWho should read it: people working on JWST high-z galaxy populations and on dark star / Pop III scenarios. It's a reasonable and mostly clear paper, but the MACHO exclusion needs a major revision. I would send it to peer review with the expectation the authors redo the comparison with proper halo growth or reframe the conclusion. Not something I'd cite yet.","headline":"A timely test of dark stars vs the JWST SFE excess, but the MACHO exclusion argument skips halo growth and overstates the relic BH abundance.","tokens_in":27137,"tokens_out":3545,"would_cite":false,"duration_ms":38289,"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":"Dark stars above ~10^3 solar masses can match JWST's star formation efficiency excess at z~11-14, but their relic black holes would overpopulate halos under current MACHO bounds, so dark stars can only be a minor contributor.","keywords":["high-redshift galaxies","dark stars","WIMP dark matter","star formation efficiency","UV luminosity function","MACHOs","Population III stars","JWST"],"falsifier":"Measure the UV luminosity function at $z\\sim13$ with high spectroscopic completeness and fit a model that lets the Pop II SFE parameters $\\epsilon_N$, $\\beta$, and $\\gamma$ evolve freely beyond $z\\sim9$; if the data are explained without any dark star component, the claimed dark star excess is not needed. Alternatively, a direct search for dark star spectral signatures in galaxies like JADES-GS-z13-0, combined with a measurement of the resulting black hole mass function in the $10^3\\!-\\!10^5\\,M_\\odot$ range, would test whether the inferred relic abundance actually exceeds MACHO bounds.","tokens_in":2132,"feed_emoji":"🌟","tokens_out":3147,"duration_ms":75604,"temperature":0.7,"pith_summary":"The paper asks whether dark stars—supersized first-generation stars fueled by WIMP annihilation instead of nuclear fusion—can explain the anomalously high star formation efficiency (SFE) inferred from JWST galaxies at redshifts 11-14. It finds that a population of dark stars with masses above roughly $10^3\\,M_\\odot$, powered by captured weakly interacting massive particles, can reproduce the observed ultraviolet luminosity function excess at these redshifts. However, the same population, if its top-heavy mass function extends to $10^4\\!-\\!10^5\\,M_\\odot$, collapses into black holes whose halo abundance would exceed current MACHO constraints from microlensing and stellar dynamics. The paper therefore concludes that dark stars can contribute only a small fraction of the excess, and that Population III stars, which have a similar UV output per unit star formation, are a more plausible candidate for the bulk.","feed_headline":"Dark stars fit JWST excess—but black holes can't hide","feed_subtitle":"WIMP-powered stars match the z≈13 UV data, but relic black holes would overpopulate galactic halos.","key_machinery":"The central machinery is the two-channel star formation efficiency decomposition, $f_{\\rm tot} = f_S + f_{DS}$, where the normal star efficiency $f_S$ is a double power law in halo mass fit at $z\\sim4{-}9$ and the dark star efficiency is a power law $f_{DS} = \\epsilon_{DS}(M_h/M_1)^{\\gamma_{DS}}$, which is used to map halo mass to ultraviolet luminosity and hence to the UV luminosity function. The second load-bearing piece is the dark star UV conversion factor $K_{\\rm UV}$, computed from a top-heavy IMF with $\\phi(m) \\propto m^{-0.17}$, which makes dark stars roughly as efficient as Pop III stars in producing UV light. The third piece is the relic black hole halo fraction, $\\psi(m) = \\epsilon_{DS,\\rm eff} f_b [\\xi_0 \\phi(m) m^2 / M_{\\rm tot}]$, which is compared with MACHO constraints from strong-lensing caustic crossings, ultra-faint dwarf heating, and gravitational scattering heat transfer.","core_discovery":"The paper argues that a population of dark stars with $M \\gtrsim 10^3\\,M_\\odot$, sustained by WIMP capture and annihilation, has a UV luminosity-to-star-formation conversion factor ($K_{\\rm UV}$) close to that of Population III stars and noticeably higher than normal Population II stars, so the JWST star formation efficiency excess at $z\\sim 11{-}14$ can be reproduced without exhausting the baryon budget. Yet the top-heavy IMF of these dark stars, with $\\phi(m) \\propto m^{+0.17}$, means their remnants become massive black holes; comparing the resulting relic black hole halo fraction with microlensing and dynamical constraints shows that a full dark star explanation is excluded unless dark star masses are confined to roughly $500\\!-\\!945\\,M_\\odot$, which supplies too little UV light. The paper's conclusion is that dark stars are not the dominant origin of the excess, while Population III stars—which share the same SFE model and a similar $K_{\\rm UV}$—remain a viable alternative.","pith_inferences":["Editorial inference: If future JWST data show that the high-redshift UV excess declines with better photometric completeness or is explained by redshift-evolving dust, IMF, or AGN contamination, the dark star (and Pop III) attribution of the residual would lose its observational basis.","Editorial inference: The same fitting framework could be used the other way around—if dark stars are confirmed, the spectrum fit favoring temperatures near $5.75\\times10^4$ K and WIMP masses from tens of GeV to a few TeV would add a new, independent probe of WIMP dark matter properties.","Editorial inference: The MACHO exclusion of very massive dark stars assumes that essentially all dark stars above $500\\,M_\\odot$ collapse to black holes with minimal mass loss; if a substantial fraction instead disrupt in pair-instability-like events or lose mass before collapse, the relic black hole abundance would be overestimated.","Editorial inference: The power-law ansatz $f_{DS} \\propto M_h^{\\gamma_{DS}}$ is an untested placeholder; zoom-in simulations of dark star formation in minihalos could directly test whether the monotonic halo-mass dependence and the fitted normalization are physical."],"forward_implications":["If dark stars with $M \\gtrsim 10^3\\,M_\\odot$ exist at $z\\sim11{-}14$, they are efficient primary UV sources and lower the star formation efficiency required per host halo.","The same population cannot have a top-heavy IMF extending to $10^4\\!-\\!10^5\\,M_\\odot$ without producing relic black holes that violate current MACHO constraints.","To survive those constraints, dark star masses must be squeezed into $500\\!-\\!945\\,M_\\odot$, a range that cannot account for the observed SFE excess.","Population III stars, with nearly the same $K_{\\rm UV}$ and the same assumed efficiency model, can mimic dark stars in UV LF fits while avoiding the MACHO overproduction problem.","The UV LF data at $z\\sim11{-}14$ require an extra component beyond the extrapolated Pop II efficiency, with the tension growing to about 2 dex at $z\\sim13$ and 3 dex at $z\\sim14$."],"supporting_citations":[{"why":"Provides the dark star H-R tracks with and without WIMP capture, from which the UV conversion factor and the mass-temperature constraints are derived.","marker":"Freese et al. (2010)"},{"why":"Supplies the UV-to-star-formation conversion factor for Pop III stars and the $\\eta_{\\rm UV}$, $\\epsilon_{\\star,\\rm rad}$ framework adopted for comparison.","marker":"Inayoshi et al. (2022)"},{"why":"Provides the UV luminosity function model mapping the halo mass function to $M_{\\rm UV}$, which the fitting procedure builds on.","marker":"Wang et al. (2023a)"},{"why":"The identification of three $\\sim10^6\\,M_\\odot$ dark star candidates that motivates testing dark stars as the source of the SFE excess.","marker":"Ilie et al. (2023)"},{"why":"Prior work estimating that a single $10^6\\,M_\\odot$ dark star cannot explain JWST UV luminosity functions, which this paper extends to a population-level analysis.","marker":"Iocco & Visinelli (2024)"},{"why":"Provides high-redshift UV luminosity function data at $z>10$ used in the fits.","marker":"Harikane et al. (2023)"},{"why":"Supplies the strong-lensing caustic-crossing microlensing constraint on the halo fraction of compact objects, used to exclude very massive dark star remnants.","marker":"Oguri et al. (2018)"},{"why":"Supplies the ultra-faint dwarf dynamical heating constraint on MACHO abundance.","marker":"Brandt (2016)"},{"why":"Supplies the gravitational-scattering heat-transfer constraint on MACHO abundance.","marker":"Graham & Ramani (2024)"},{"why":"Provides the spectrum of the $z=13.2$ galaxy JADES-GS-z13-0 used to constrain dark star temperature and the dark star fraction of the UV radiation.","marker":"Curtis-Lake et al. (2023)"}],"fun_headline_variants":["Dark stars fail JWST excess test—black hole relics too many","JWST excess stays unexplained—dark stars ruled out","Dark stars can't lift JWST excess—remnants would flood halos","Population III stars beat dark stars for JWST excess","Dark star black holes kill the JWST excess fix"],"cache_read_input_tokens":29056,"weakest_assumption_plain":"The entire residual UV luminosity at $z\\sim11{-}14$ is attributed to dark stars (or Pop III stars) through a power-law efficiency while ordinary Pop II star formation efficiency is assumed to stop evolving beyond $z\\sim9$; if the Pop II efficiency continues to rise with redshift, or the residual comes from dust, IMF variation, or AGN contamination, the fitted dark star fraction and the MACHO exclusion built on it would not follow.","fun_headline_variants_meta":{"raw":{"variants":["Dark stars fail JWST excess test—black hole relics too many","JWST excess stays unexplained—dark stars ruled out","Dark stars can't lift JWST excess—remnants would flood halos","Population III stars beat dark stars for JWST excess","Dark star black holes kill the JWST excess fix"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000776,"raw_usage":{"total_tokens":3457,"prompt_tokens":992,"completion_tokens":2465,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":608,"completion_tokens_details":{"reasoning_tokens":2381}},"tokens_in":608,"tokens_out":2465,"duration_ms":18004,"temperature":1.0,"reasoning_tokens":2381,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-10T20:49:59.545323+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Measure the UV luminosity function at $z\\sim13$ with high spectroscopic completeness and fit a model that lets the Pop II SFE parameters $\\epsilon_N$, $\\beta$, and $\\gamma$ evolve freely beyond $z\\sim9$; if the data are explained without any dark star component, the claimed dark star excess is not needed. Alternatively, a direct search for dark star spectral signatures in galaxies like JADES-GS-z13-0, combined with a measurement of the resulting black hole mass function in the $10^3\\!-\\!10^5\\,M_\\odot$ range, would test whether the inferred relic abundance actually exceeds MACHO bounds.","supporting_citations":[{"cited_title":"W., & Ramani , H","cited_arxiv_id":null,"evidence_quote":"Supplies the gravitational-scattering heat-transfer constraint on MACHO abundance."}],"review_version":1}