{"id":"0c57c29d-22fa-4afc-bfe8-c3756fb48afa","arxiv_id":"2502.04432","paper_version":2,"verdict":"CONDITIONAL","confidence":"HIGH","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":5,"one_line_summary":"Hyperlight quadratically coupled scalars making up a few percent of dark matter are bounded to near- or sub-gravitational couplings to electrons and photons over 10^-31 to 10^-28 eV, with quasar spectra setting the stronger limits below 10^-31 eV.","lead":"This paper maps the allowed couplings of extremely light, cosmologically frozen scalar fields that would subtly change the electron mass and the strength of electromagnetism over cosmic history. It combines early-universe data (CMB, baryon oscillations, supernovae, nucleosynthesis) with quasar and laboratory searches to rule out couplings down to near gravitational strength for these fields.","discovery_kind":"new_application","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Electron-mass exclusion is dataset-dependent: DESI yields a ~2σ positive coupling, so the 'most stringent' claim is not robust.","rationale":"The reader's weakest_assumption (frozen scalar) is explicitly bounded by Eq. (4.15), and the reported parameter space satisfies |d^(2)| ≲ 500, so it is unlikely to change the exclusion limits at the quoted precision. The quasar posteriors are improper only individually; the joint posteriors with N≥14 fall off as d^{-N/2} and are proper, so that concern is less load-bearing. The most load-bearing concern is the dataset dependence of the electron-mass constraint, which directly affects the 'most stringent' claim for one of the two couplings. The paper itself shows that CMB+DESI prefers a positive electron-mass shift at ~2σ, while other datasets are consistent with zero; this is not an issue of scalar dynamics but of the data used to break the CMB degeneracy. If the DESI preference is real, the paper would be reporting a hint rather than a constraint; if it is a systematic, the constraints are not stable. The proposed check—a joint fit with systematics—would settle this. Because the paper discloses the discordance and the central framework is otherwise sound, the CONDITIONAL verdict remains appropriate.","tokens_in":56414,"tokens_out":14653,"duration_ms":153381,"concrete_test":"Recompute the joint constraint on 10^2(me,i/me,0−1) using Planck 2018 plus a combined low-redshift likelihood that includes DESI, SDSS+6dFGS, DES, and Pantheon+ with a nuisance offset parameter per sample (or a Bayesian hierarchical model for distance-scale systematics). If the resulting 95% interval is consistent with zero and stable under variations of the nuisance treatment, the discordance is resolved and the 'most stringent' claim holds; if the interval remains nonzero or shifts by more than the quoted 1σ uncertainties, the electron-coupling constraint is dataset-dependent and the abstract should be qualified.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central claim that CMB+BAO+SNe provide the most stringent constraints is not robust for the quadratic electron coupling. Section IV B reports that the inferred early-time electron mass depends strongly on the low-redshift dataset: Planck+DESI gives 10^2(me,i/me,0−1)=1.7±0.7, while Planck+DES gives −1.6±1, Planck+SDSS gives 0.8±0.7, and Planck+Pantheon+ gives −0.6±1.1. Figure 5 translates this into a ~2σ preference for positive d^(2)_me from CMB+DESI, whereas other combinations are consistent with zero. Thus the abstract's unqualified 'most stringent constraints' on quadratically coupled scalars is only valid for a particular choice of low-redshift data; for positive electron couplings, the DESI combination does not exclude d=0 at 95% CL. The paper discloses this discordance, but it is exactly the dataset-dependence that determines whether the exclusion map in Fig. 5 stands as the headline result.","agreement_with_reader":"partial"},"referee_report":{"model":"deepseek-v4-flash","summary":"This paper studies hyperlight scalar fields coupled quadratically to the electron mass and the fine-structure constant, with masses in the range roughly 10^-32 to 10^-28 eV and present-day abundances of order a percent of the cold dark matter density. The authors compute the in-medium effective potential sourced by Standard Model particles, identify when the scalar can remain frozen through recombination, extend late-time probes (quasars, atomic clocks, equivalence-principle tests, Oklo, pulsar timing, stellar emission) to this regime, and combine these with early-time BBN and CMB+BAO+SNe bounds. The main result is an exclusion map for the quadratic couplings d^(2)_me and d^(2)_e, with the CMB-based analysis (imported from the companion paper [60]) providing the most stringent constraints for masses from about 10^-28.5 eV down to 10^-31 eV, below which quasar absorption spectra become more powerful.","tokens_in":56553,"tokens_out":6672,"duration_ms":75241,"significance":"If the central claim holds, this is the most complete current map of where hyperlight, quadratically coupled scalars can live, and it substantially sharpens earlier phenomenological bounds by grounding them in a concrete field-theoretic model with explicit cosmological dynamics. The paper's strengths are its careful treatment of in-medium matter potentials (with analytic cross-checks in Appendix A), its explicit delineation of the regime where the scalar is frozen through recombination, and its honest reporting of dataset-dependent results in the electron-mass channel. The companion CMB likelihood analysis [60] is separately published and data-driven, so importing its constraints is a reasonable division of labor rather than a circular step. However, the headline claim of 'most stringent constraints' is not uniformly robust: the electron-mass coupling exclusions depend strongly on the choice of low-redshift dataset, as the paper itself documents, and the quasar electron-mass bound at the lowest masses relies on phase-marginalization assumptions that the authors only partially validate.","major_comments":[{"comment":"The abstract's unqualified statement that the CMB+BAO+SNe analysis 'provides the most stringent constraints' is not robust for the electron-mass coupling. Section IV B reports 10^2(me,i/me,0 - 1) = 1.7 +/- 0.7 for Planck+DESI, -1.6 +/- 1 for Planck+DES, 0.8 +/- 0.7 for Planck+SDSS, and -0.6 +/- 1.1 for Planck+Pantheon+, and Figure 5 shows a roughly 2-sigma preference for positive d^(2)_me for the DESI combination while other combinations are consistent with zero. Thus for positive electron couplings, the DESI-based bound does not exclude zero at 95% CL, and the 'most stringent' claim holds only for a particular dataset choice. Although the text discloses the discordance, the abstract and Section IV C present the result without this qualification. The authors should either report the dataset-dependent range as the headline electron-coupling result or justify a specific conservative combination.","section":"Section IV B and Figure 5"},{"comment":"The quasar-derived electron-mass constraint that supersedes the CMB bounds at the lowest masses is not as robust as the photon-coupling result. The phase-marginalized posterior in Figure 3 is visibly skewed toward negative Delta me,0/me,0, and the text states that this skew is driven by a cluster of higher-redshift observations and that the joint bound is inappropriate for m_phi below about 10^-31 eV, where the z_abs < 1 subset gives a much weaker bound. Since the abstract explicitly claims quasar absorption spectra provide stronger bounds below 10^-31 eV, the electron-coupling branch of that claim depends on a set of high-redshift absorbers and an independent-phase assumption that the authors themselves flag as potentially obscuring correlations. This part of the headline should be qualified, or the analysis should be redone with a more careful treatment of phase correlations across the absorber sample.","section":"Section III C and Eq. (3.8)"},{"comment":"The paper's presentation of the CMB-based constraints on d^(2)_me and d^(2)_e mixes likelihood-driven results with prior-driven ones. Appendix B shows that changing from uniform priors over F_phi and the early-time parameter value to a uniform prior over d^(2)_lambda and a log-uniform prior over the initial field value broadens the marginalized posterior over d^(2)_me by about an order of magnitude, even though the posterior over me,i itself is relatively stable. Since Figure 5 and the abstract present d^(2)_lambda exclusions at fixed F_phi = 10^-2, the authors should state more prominently that the marginalized constraints in Figure 4 are prior-dependent and that the robust, likelihood-driven quantity is the early-time shift combination d^(2)_lambda F_phi, not d^(2)_lambda itself.","section":"Section IV C and Appendix B"}],"minor_comments":[{"comment":"The BBN constraints are derived from a simplified two-step freeze-out calculation rather than a full nuclear network, and the paper acknowledges that this calculation is insufficient to predict Y_p accurately. Since the CMB bounds are stronger in most of the displayed mass range, this is not fatal, but the authors should either validate Eq. (4.10) against a public BBN code or explicitly state that the BBN curves in Figure 5 are indicative only.","section":"Section IV A, Eq. (4.10)"},{"comment":"The parameter D defined in Eq. (2.49) combines electron and photon couplings through the weighted sum of dilatonic charges, while later conditions such as Eq. (4.15) are written for individual couplings. The connection between these two levels of approximation could be stated more explicitly to avoid confusion about when matter effects are controlled by |D| versus by |d^(2)_lambda|.","section":"Section II E 2, Eq. (2.49)"},{"comment":"The quasar sample contains a few extremely large uncertainties, especially the J1120+0641 measurements at z ~ 5-7 with errors of order 10^-4 in Delta alpha/alpha. The text discusses redshift subsamples but does not state how these particular high-redshift, high-uncertainty points affect the joint posterior; a sentence quantifying their influence would make the phase-marginalization check more transparent.","section":"Section III C, Table I"},{"comment":"The sentence 'DES, SDSS, and Pantheon datasets rule out |d^(2)_me| smaller than a few' is ambiguous, because the sign of the coupling matters and the displayed posteriors show asymmetric bounds. It would be clearer to quote the actual 95% intervals for each dataset combination.","section":"Section V C"}],"recommendation":"major_revision","confidential_remarks":"The core calculation is careful and the companion analysis [60] gives the paper a solid empirical anchor, but the abstract's 'most stringent constraints' claim overstates the robustness of the electron-mass channel. The dataset-dependence is documented in the paper itself, so the fix is wording and presentation rather than new physics, but it is a central part of the headline result. The quasar electron-mass bound at low masses also deserves a more cautious statement. I do not see a circularity problem with the use of [60]."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"This paper deserves a serious referee. It's a constraints paper with a genuinely new exclusion map for hyperlight, quadratically coupled scalars in the 10^-32 to 10^-28 eV mass range, and it is unusually good about saying where its own assumptions break.\n\nThe genuinely new content is the systematic computation of in-medium potentials across cosmic history, the phase-marginalized reinterpretation of quasar and laboratory searches, and the joint early/late Universe exclusion map in Fig. 5. The in-medium potential calculation is careful, analytically cross-checked in Appendix A, and the validity regimes (frozen scalar through recombination, coupling ceiling |d| ~ 500) are stated explicitly. That is real work, and it gives future experiments a target map.\n\nThe main soft spot is the one the stress-test flags: the electron-mass constraint is dataset-dependent. Planck+DESI prefers a positive coupling at ~2 sigma, while Planck+DES, Planck+SDSS, and Planck+Pantheon+ are all consistent with zero. So the abstract's 'most stringent constraints' is only robust for the photon coupling; for the electron coupling it depends on which low-z dataset you pick. To the paper's credit, Section IV.B and Fig. 5 present all combinations and openly discuss the discordance. But the abstract oversells it, and that should be fixed before publication.\n\nThe quasar constraints rest on improper posteriors that need an ad hoc phase regularization; the authors disclose this and argue the result is conservative, but it is a genuine caveat. The BBN bound uses an approximate Y_p scaling that the authors explicitly flag. The central CMB machinery is imported from the companion paper [60], which is published and data-driven, so not circular, though an independent re-implementation would be needed to fully trust the map. None of these are fatal; they are the usual gaps in a broad constraints paper.\n\nOverall: the central exclusion map is credible, the theoretical framework is sound, and the paper is unusually honest about its limitations. The dataset-dependence of the electron bound should be reflected in the abstract, and the quasar regularization deserves a bit more discussion. I would bring this to a reading group and would cite it for the in-medium potential and the exclusion map. Send it to peer review; a competent referee will want minor-to-moderate revisions, not a reject.","headline":"A careful, unusually candid constraints paper that maps where hyperlight quadratically coupled scalars can live; the headline CMB+BAO/SNe bounds for electron couplings are dataset-dependent, but the paper says so.","tokens_in":57215,"tokens_out":1367,"would_cite":true,"duration_ms":15777,"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":"The paper claims that early-universe cosmology, not laboratories, sets the strongest limits on quadratically coupled hyperlight scalars, restricting percent-level dark-matter subcomponents to near- or sub-gravitational couplings to…","keywords":["hyperlight scalar","quadratic coupling","fine-structure constant","electron mass","cosmic microwave background","quasar absorption lines","dark matter subcomponent","varying fundamental constants"],"falsifier":"Recompute the CMB damping-tail and polarization bounds with a scalar that is allowed to roll during recombination, taking $|D| > 1$ at couplings $d_\\lambda^{(2)} \\approx 500$ and abundance $F_\\varphi \\approx 10^{-4}$; if the resulting exclusions on $d_\\lambda^{(2)}F_\\varphi$ differ from Eq. (2.57), the reported exclusion map must be revised in that corner.","tokens_in":56082,"feed_emoji":"🌌","tokens_out":9849,"duration_ms":97987,"temperature":0.7,"pith_summary":"The paper tries to establish where a new class of hyperlight scalar fields—particles with masses around $10^{-32}$ to $10^{-28}$ eV that are quadratically coupled to the electron mass or the fine-structure constant—can still exist. Because such a scalar is frozen early in cosmic history and only begins oscillating near the present, its main signature is a constant shift in $\\alpha$ and $m_e$ during BBN and recombination, together with a gravitational contribution to dark matter. The authors compute how the Standard Model bath back-reacts on the scalar through an in-medium mass, which confines viable models to a narrow coupling window. Within that window they combine BBN, CMB, BAO, and supernova constraints with quasar absorption spectra, concluding that CMB plus BAO plus supernova data give the strongest bounds for masses from $10^{-28.5}$ to about $10^{-31}$ eV, below which quasar spectra dominate. The net result is that a scalar making up a few percent of today's dark matter must couple to electrons or photons at near- or sub-gravitational strength.","feed_headline":"CMB and quasars box in hyperlight scalar dark matter","feed_subtitle":"If such scalars are 1% of dark matter, electron and photon couplings sit near or below gravitational strength.","key_machinery":"The argument is carried by the quadratic coupling functions $g_\\lambda(\\varphi) = d_\\lambda^{(2)}\\varphi^2/2$, which make the electron mass and fine-structure constant depend on the scalar field. The load-bearing mechanism is the in-medium, thermal mass that the Standard Model bath induces on the scalar, parametrized after electron-positron annihilation by the dimensionless combination $D = (3/2)(Q_b)_\\lambda d_\\lambda^{(2)}/(1+\\omega_c/\\omega_b)$. Requiring $|D| \\lesssim 1$ keeps the scalar frozen through recombination; the paper derives analytic solutions (Bessel-type before equality, power-law during matter domination) showing that for $|D| \\gtrsim 1$ the field rolls before last scattering, invalidating the frozen-field mapping. The mapping itself, Eq. (2.57), converts early-time constant shifts into the product $d_\\lambda^{(2)}F_\\varphi$, and the cosmological data from Ref. [60] bound $F_\\varphi$ and the shifts jointly. For late-time probes, the long oscillation period makes the phase of the scalar unknown, so bounds are obtained by marginalizing over phase, which strongly penalizes atomic clock and equivalence-principle constraints.","core_discovery":"The central discovery is that the time-independent shifts in $\\alpha$ and $m_e$ that prior phenomenological studies assumed at recombination are only realizable by a scalar in a precisely delimited regime: it must be heavy enough to be frozen until after last scattering, light enough to be relevant today, and coupled weakly enough that matter-induced potentials do not make it roll before recombination ($|D| \\lesssim 1$, $|d_\\lambda^{(2)}| \\lesssim 500$), yet strongly enough that its early-time shift is observable. In this regime the early-time shift is set by Eq. (2.57), $\\Delta\\lambda_i/\\lambda_0 \\approx (2/3)d_\\lambda^{(2)}F_\\varphi/(1+\\omega_b/\\omega_c)$, so a percent-level abundance $F_\\varphi$ maps directly to an allowed coupling. Using the companion CMB analysis (Ref. [60]), the paper reports the most stringent constraints on quadratically coupled scalars with masses $10^{-28.5}$ to $10^{-31}$ eV, with quasar absorption spectra taking over below about $10^{-31}$ eV. For the electron coupling the bound is dataset-dependent, with BAO and supernova combinations disagreeing at the one-to-three $\\sigma$ level, while the photon-coupling bound is robust. The result is an exclusion map in which hyperlight scalars that make up a few percent of dark matter are limited to near- or sub-gravitational quadratic couplings to electrons or photons.","pith_inferences":["If the exclusion map is taken at face value, any model that uses a quadratically coupled scalar to shift $\\alpha$ or $m_e$ at recombination, for example to ease the Hubble tension, must have $F_\\varphi$ near the percent level and couplings below roughly gravitational strength, otherwise it is already ruled out.","The asymmetry between positive and negative couplings, with negative couplings producing growing rather than oscillating solutions, suggests that a dedicated dynamical analysis for negative $d_\\lambda^{(2)}$, beyond the frozen-field approximation, could close the remaining window at $F_\\varphi \\lesssim 10^{-4}$.","Future galaxy-survey and CMB data should sharpen the electron-coupling bound and resolve the current BAO and supernova discordance; if the discordance persists, the constraint on $d_{m_e}^{(2)}$ will be limited by systematics rather than by the scalar model itself."],"forward_implications":["If the scalar is frozen through recombination, early-universe data (CMB, BAO, supernovae) exclude quadratic couplings to photons and electrons far more strongly than any terrestrial or astrophysical late-time probe, except at the very lightest masses.","A scalar that makes up roughly one to two percent of the dark matter today can no longer hide: at masses $10^{-28.5}$ to $10^{-31}$ eV its couplings to electrons or photons must be near or below gravitational strength.","Quasar absorption spectra remain the leading probe below about $10^{-31}$ eV, because such scalars stay frozen and their effect on $\\alpha$ and $m_e$ grows as $(1+z)^3$; single finite-duration clock experiments cannot compete because phase marginalization penalizes them.","Existing phenomenological bounds on time-independent shifts at recombination are only valid for a nonnegligible scalar abundance, roughly $F_\\varphi \\gtrsim 10^{-4}$ to $10^{-5}$; below that, in-medium dynamics during recombination require a fully time-dependent treatment.","The electron-coupling limit from cosmological data is currently dataset-dependent, with BAO and supernova combinations differing, so the reported exclusion for $d_{m_e}^{(2)}$ should be read as a range, while the photon-coupling limit is robust."],"supporting_citations":[{"why":"Supplies the companion CMB, BAO, and supernova analysis whose posteriors on $\\alpha_i$, $m_{e,i}$, and $F_\\varphi$ yield the paper's leading constraints in the $10^{-28.5}$ to $10^{-31}$ eV window.","marker":"[60]"},{"why":"Provides the BBN bounds on quadratically coupled ultralight scalar dark matter and the WKB and in-medium dynamics that the paper extends to hyperlight masses.","marker":"[37]"},{"why":"Derives BBN constraints for universally coupled ultralight scalars from the in-medium potential, a key input to the early-time analysis.","marker":"[59]"},{"why":"Defines the dilatonic charges mapping fundamental couplings to composite matter, used to compute matter potentials and equivalence-principle signals.","marker":"[21]"},{"why":"Companion work supplying the charge matrix and coupling-to-composite relations used for nuclei, atoms, and the Earth.","marker":"[22]"},{"why":"Provides the equivalence-principle framework and screening profile for quadratically coupled scalar dark matter, used for the late-time recast.","marker":"[26]"},{"why":"Compiles the molecular hydrogen quasar absorption measurements of $\\Delta\\mu/\\mu$ that set the light-mass electron-coupling constraints.","marker":"[118]"},{"why":"Supplies the zinc and chromium quasar absorption measurements of $\\Delta\\alpha/\\alpha$ that set the light-mass photon-coupling constraints.","marker":"[127]"}],"fun_headline_variants":["Hyperlight scalars squeezed to near-gravitational couplings","CMB, quasars, and supernovae box in scalar dark matter","Early Universe limits hyperlight scalar couplings to gravity strength","Hyperlight scalar dark matter: only weak couplings pass cosmic tests","Most stringent bounds yet on hyperlight scalar dark matter"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The load-bearing assumption is that the scalar is effectively frozen and time-independent through recombination, so CMB and BBN probes see constant but shifted values of $\\alpha$ and $m_e$; the paper bounds this assumption by requiring $|D| \\lesssim 1$ and $|d_\\lambda^{(2)}| \\lesssim 500$, but if a scalar were dynamical during recombination the reported coupling limits would need to be re-derived.","fun_headline_variants_meta":{"raw":{"variants":["Hyperlight scalars squeezed to near-gravitational couplings","CMB, quasars, and supernovae box in scalar dark matter","Early Universe limits hyperlight scalar couplings to gravity strength","Hyperlight scalar dark matter: only weak couplings pass cosmic tests","Most stringent bounds yet on hyperlight scalar dark matter"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000423,"raw_usage":{"total_tokens":2253,"prompt_tokens":1105,"completion_tokens":1148,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":721,"completion_tokens_details":{"reasoning_tokens":1064}},"tokens_in":721,"tokens_out":1148,"duration_ms":11569,"temperature":1.0,"reasoning_tokens":1064,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-08T22:43:30.388916+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Recompute the CMB damping-tail and polarization bounds with a scalar that is allowed to roll during recombination, taking $|D| > 1$ at couplings $d_\\lambda^{(2)} \\approx 500$ and abundance $F_\\varphi \\approx 10^{-4}$; if the resulting exclusions on $d_\\lambda^{(2)}F_\\varphi$ differ from Eq. (2.57), the reported exclusion map must be revised in that corner.","supporting_citations":[],"review_version":1}