{"id":"2835c05a-c713-461a-a958-dc274be4b15b","arxiv_id":"2608.11620","paper_version":1,"verdict":"ACCEPT","confidence":"HIGH","novelty_score":4.0,"correctness_risk":"low","formal_verification":"none","parameter_count":3,"one_line_summary":"No gravitational-wave signal was found from boson clouds around two black hole merger remnants, and three high-spin binaries exclude scalar and vector boson masses near 10^-13 eV at 90 percent confidence.","lead":"The LIGO-Virgo-KAGRA collaboration searched for gravitational waves from hypothetical ultralight boson clouds around two newly formed black holes, heard nothing, and now disfavor a narrow slice of the allowed boson mass range with high confidence. Spin measurements of three other black hole binaries independently exclude overlapping masses, sharpening the experimental hunt for a class of dark matter candidates.","discovery_kind":"extension","skeptic_critique":{"model":"deepseek-v4-flash","headline":"SuperRad model fidelity is the load-bearing input for the quoted 90% exclusion range; no independent validation is provided in this paper.","rationale":"The reader's verdict ACCEPT with HIGH confidence is defensible. The paper is careful: it says 'disfavored' rather than 'excluded,' acknowledges the Gaussian-noise approximation, and notes that line-affected bands may have optimistic Pdet. My stress-test agrees with the reader's weakest-assumption identification: the SuperRad model is the load-bearing input. However, identifying a load-bearing assumption is not the same as finding a flaw. The paper does not hide this dependence; it is stated in Sec. IIIB and Appendix A. Still, because the headline number—'>90% confidence' for a specific mass interval—is a direct output of SuperRad's predictions, a factor-of-few error in the model's saturation energy or timescales would change the headline. A single independent cross-check would settle whether this is a practical concern or only a formal one. Given the transparency and the standard practice in this field, I do not think the verdict should change; UNCHANGED is appropriate. The Gaussian-noise sensitivity estimation is a secondary concern, but it is subordinate to the signal-model dependence.","tokens_in":43397,"tokens_out":29710,"duration_ms":274918,"concrete_test":"Use an independent code (e.g., the Baryakhtar et al. (2017) vector-cloud formalism or a published numerical-relativity calibration) to compute h0, tau_growth, tau_GW, and f_dot for the median remnant parameters of GW250114 and GW250207. Compare with Table III. Then re-run the Sec. IIIE injection pipeline for mV in [2.5, 4.0] x 10^-13 eV using the alternative waveform, or with SuperRad h0 scaled by 0.5 and 2.0, and recompute Pdet(mV). If Pdet falls below 0.9 anywhere in [2.80, 3.95] x 10^-13 eV, the headline confidence statement is not robust to model uncertainty; if Pdet remains above 0.9, the concern is resolved.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central quantitative claim—that vector boson masses in [2.80, 3.95] x 10^-13 eV are disfavored at >90% confidence—rests entirely on the SuperRad waveform model. SuperRad sets the signal parameters used both to configure the HMM search (tau_growth, tau_GW, t_start, T_coh via f_dot; Sec. IIIB, Table III, Sec. IIIC) and to compute the detection probability Pdet(mV) in the injection simulations (Sec. IIIE, Eq. 2). The same model supplies chi_max for the spin-based exclusion (Sec. IVA). If SuperRad's predicted strain amplitude h0 is too high by a factor of ~2, or if the true cloud saturates later or depletes faster, then Pdet(mV) would be overestimated and the 90% confidence interval would shrink or shift. The paper does not provide a validation of SuperRad for these specific remnant parameters (M ~ 62.7 M_sun, chi ~ 0.68), nor any uncertainty quantification on the model outputs. This is not an internal inconsistency—the paper is explicit that the ranges are sensitivity estimates under a signal model—but it is the least secure link in the causal chain from 'no detection' to 'masses disfavored at 90% confidence.' The spin-based results inherit the same model dependence through chi_max, making the concern compound.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper presents constraints on ultralight bosons using data from the second and third parts of the fourth LIGO-Virgo-KAGRA observing run (O4b/O4c), with two complementary analyses. The first is a directed hidden-Markov-model search for long-transient gravitational waves from vector boson clouds around the merger remnants of GW250114 and GW250207; no candidate signals survive follow-up, and the authors quote a disfavored vector boson mass range of [2.80, 3.95] x 10^-13 eV at greater than 90% confidence for a 1% false-alarm threshold. The second analysis uses the high spins of the pre-merger black holes in GW240515, GW241113, and GW241225_08 to exclude, at 10^5 yr assumed black-hole age, scalar masses in [1.39, 6.94] x 10^-13 eV and vector masses in [0.32, 14.4] x 10^-13 eV at 90% confidence. The paper is explicit that the directed-search ranges are sensitivity estimates under a signal model and that the spin-based ranges depend on assumed black-hole age.","tokens_in":43628,"tokens_out":10094,"duration_ms":117486,"significance":"If the central claims hold, the paper provides the strongest directed-search constraints to date on vector boson clouds around recent merger remnants and extends spin-based superradiance constraints to new O4b events. The analysis is notably careful in several respects: it labels the remnant-search result as 'disfavored' rather than 'excluded' because sensitivity is estimated with matched Gaussian noise; it quotes the combined interval as a union rather than a joint combination; it uses a prior-driven non-exclusion criterion in Eq. (4) to ensure that spin-based exclusions are data-driven; and it documents the candidate veto chain and search configurations in detail. The injection-based detection probability, Eq. (2), and the explicit percentile-based search configurations provide a reproducible framework. The main risk, as the authors themselves acknowledge, is that both the directed-search sensitivity and the spin-based chi_max are computed with the SuperRad signal model without an independent validation or uncertainty quantification for the specific remnant and binary parameters considered.","major_comments":[{"comment":"Pdet(mV) is computed by injecting synthetic signals into matched Gaussian noise and applying only the threshold-based candidate selection; the follow-up veto chain is not applied to the injections. The search itself vetoed all surviving candidates for GW250114 and ten candidates for GW250207 as instrumental artifacts, so on real data the detection efficiency is potentially lower than the quoted Pdet. Because the abstract's central interval is defined by Pdet >= 0.9, the >90% confidence claim is not directly established on real-data terms; please either apply the full veto chain to injected signals, inject into representative off-source real noise, or provide a quantitative estimate of the resulting sensitivity loss and adjust the confidence statement accordingly.","section":"Sec. IIIE, Eq. (2)"},{"comment":"The signal parameters that set the search configuration and the injection-based sensitivity (tau_growth, tau_GW, h0, f0, fdot) are taken solely from the SuperRad waveform model, with no independent validation or uncertainty quantification for the specific remnant parameters (M ~ 62.7 M_sun, chi ~ 0.68) of GW250114 and GW250207. Since both the directed-search disfavored range and the spin-based chi_max in Sec. IVA inherit this model dependence, a factor-of-two change in the predicted strain amplitude or growth/depletion timescales could shift or shrink the quoted intervals. The use of the word 'disfavored' tempers the claim, but the manuscript should either cross-check SuperRad against an independent calculation or perform a parameter-uncertainty scan showing that the quoted mass ranges are robust to plausible model variations.","section":"Sec. IIIB, Table III and Appendix A"}],"minor_comments":[{"comment":"The distinction between 'disfavored' for the directed search and 'excluded' for the spin-based analysis, both quoted at 90% confidence, is easy to misread; please add one sentence stating explicitly that the former is a sensitivity-based statement under a signal model while the latter is a data-driven exclusion under an assumed black-hole age.","section":"Abstract and Sec. IIIE"},{"comment":"The reported uncertainty on cos(iota) for GW250114 extends below -1 as written (0.64+0.18-1.4); since cos(iota) is bounded, consider reporting the 90% credible interval instead of median with asymmetric errors to avoid an apparently unphysical bound.","section":"Table I"},{"comment":"The spin-based exclusions are quoted at Tage = 10^5 yr, but the dependence on Tage is only shown in Fig. 4; consider explicitly stating in the abstract or conclusions that the ranges broaden with increasing assumed black-hole age and are therefore not universal.","section":"Sec. IVA"}],"recommendation":"major_revision","confidential_remarks":"The main risk is the combined model dependence: SuperRad is developed largely within the same collaboration, and the paper provides no independent validation of its strain amplitude, growth, and depletion timescales for these remnant parameters, while the injection-based sensitivity estimate also omits the follow-up veto chain. These are fixable through additional robustness checks or by softening the confidence statements. If the authors address these points, the paper is likely acceptable."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Two things to know. First, the genuinely new content is the >90% vector-boson exclusion near 3e-13 eV from the GW250114 and GW250207 remnants, plus the scalar and vector exclusions from three high-spin O4b binaries. The methods are inherited: the HMM directed search and the spin-based exclusion were developed and used in earlier papers. Second, the headline claim is conditional on the SuperRad signal model. The paper says this clearly, but it is the load-bearing point for any reader.\n\nI agree with the reader's accept verdict. The analysis is careful in the ways that matter. The authors label the remnant result as \"disfavored\" rather than \"excluded\" because sensitivity is estimated on matched Gaussian noise, and they explicitly note that line-affected frequency bands may make detection probabilities optimistic. The union of the two event ranges rather than a joint combination is the right call. The spin-based analysis includes a prior-driven non-exclusion criterion, which guards against spin-prior artifacts, and the quoted ranges are shown as functions of assumed black-hole age. Candidate follow-up is thorough: known-line vetoes, single-interferometer checks, and manual spectrogram inspection. Citation practice looks fine, with the relevant method papers and prior O4a analysis cited.\n\nThe stress-test note lands. SuperRad is load-bearing in both channels. It sets the HMM search configurations through growth and depletion timescales, the frequency drift, and the strain amplitude; it also supplies chi_max for the spin exclusions. The paper does not provide an independent validation of SuperRad for these specific remnant parameters, nor a robustness sweep over its key outputs. This is not an internal inconsistency, and the authors do disclose the model dependence, but a referee should ask for an explicit sensitivity check: for example, how much do the quoted intervals change if h0 is reduced by a factor of two, or if the saturation time shifts? That is a revision request, not a rejection.\n\nThe remaining soft spots are minor and mostly self-reported. The Gaussian-noise sensitivity estimate makes the 90% confidence a statement about the pipeline under idealized noise rather than the real data; the authors chose the word \"disfavored\" for exactly this reason. The spin-based exclusions depend on assumed black-hole ages, and the paper presents that dependence transparently.\n\nWho is this for? Anyone working on ultralight boson constraints from gravitational waves, and more broadly anyone who wants a current example of how LVK null searches are turned into exclusion statements with honest caveats. It deserves a serious referee, not a desk reject. My recommendation is to send it to review with the expectation of a moderate revision focused on model robustness.\n\nI would cite it, and I would probably bring it to a reading group to discuss how much of the 90% confidence is model rather than data.","headline":"A solid, honest constraints paper: new >90% remnant-based exclusion, but the quoted confidence is SuperRad-conditional and should be peer-reviewed with that caveat front and center.","tokens_in":54803,"tokens_out":2354,"would_cite":true,"duration_ms":31637,"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 reports that searches for long-transient gravitational waves from ultralight vector boson clouds around the remnant black holes of GW250114 and GW250207 found no signal, disfavoring vector boson masses in [2.80, 3.95] × 10^-13…","keywords":["ultralight bosons","superradiance","gravitational waves","black hole remnants","hidden Markov model","vector bosons","black hole spin","fourth observing run"],"falsifier":"A detected long-transient signal from either remnant at the frequency and drift predicted by SuperRad for a vector mass in the disfavored range would refute the no-signal claim, as would an independent analysis that recovers such a signal in the same data. Conversely, recomputing the sensitivity with an alternative waveform model that drops the detection probability below 90% would show that the quoted range is a modeling artifact rather than an established absence.","tokens_in":43188,"feed_emoji":"🕳️","tokens_out":9083,"duration_ms":93315,"temperature":0.7,"pith_summary":"This paper tests whether ultralight bosons form detectable superradiant clouds around black holes using gravitational-wave data from the second and third parts of the fourth observing run. Directed searches for long, frequency-drifting gravitational waves from vector boson clouds around the remnant black holes of GW250114 and GW250207 find no signal, disfavoring vector masses $[2.80, 3.95]\\times10^{-13}$ eV at more than 90% confidence at a 1% false-alarm threshold. A complementary spin-based analysis of three high-spin binaries excludes scalar masses $[1.39, 6.94]\\times10^{-13}$ eV and vector masses $[0.32, 14.4]\\times10^{-13}$ eV at 90% confidence for black-hole ages of $10^5$ years. If correct, these results tighten the allowed mass window for ultralight bosons around $10^{-13}$ eV, the range in which they are often proposed as dark matter or as solutions to the strong CP problem.","feed_headline":"Two mergers disfavor ultralight vector bosons at 90% confidence","feed_subtitle":"Gravitational-wave searches of two remnants and spins of three binaries narrow the allowed boson mass window.","key_machinery":"The driving mechanism is the superradiant instability: a rotating black hole spinning fast enough to satisfy $\\omega_b < m\\Omega_H$ amplifies an ultralight bosonic bound state, extracting rotational energy until the cloud saturates and then radiates nearly monochromatic gravitational waves with a small positive frequency drift. The search side is carried by the named waveform model SuperRad, which predicts the strain amplitude, growth time, depletion time, and frequency evolution that set the search configurations and the injected-signal sensitivity; the hidden-Markov-model tracker then follows the drifting signal across short Fourier-transform segments. The spin side is carried by the computed maximum spin $\\chi_{\\max}(M, m_b, T_{\\rm age})$ after superradiant spin-down, which turns a single high-spin measurement into an exclusion region in the boson-mass–black-hole-mass plane.","core_discovery":"The paper's central claim is that no vector-boson cloud emission above the estimated sensitivity exists around the two merger remnants, so vector bosons with masses $[2.80, 3.95]\\times10^{-13}$ eV are disfavored with greater than 90% confidence at a 1% false-alarm probability. The companion claim is that the high measured spins of the primary black holes in GW240515, GW241113, and GW241225_08 are inconsistent with efficient superradiant spin-down by scalar bosons in $[1.39, 6.94]\\times10^{-13}$ eV and by vector bosons in $[0.32, 14.4]\\times10^{-13}$ eV, assuming black-hole ages of $10^5$ years. The two approaches are complementary: the remnant search makes minimal assumptions about binary evolution and targets freshly formed clouds, while the spin analysis probes long-term spin-down and covers a wider mass range.","pith_inferences":["Applying the same directed-search pipeline to older remnant candidates should extend the disfavored range to lower masses, since longer-lived clouds allow longer coherent integrations.","The quoted remnant constraints are tied to SuperRad's predicted strain-versus-mass relation; recomputing the sensitivity with a different nonlinear cloud-evolution model would shift the disfavored interval and reveal how much of the result is model-driven.","Combining the spin-based exclusions with independent age estimates for dynamically formed binaries could turn the assumed $10^5$-year age into a measured prior and sharpen the mass ranges.","A single future observation of a black hole older than $10^5$ years with spin above the predicted $\\chi_{\\max}$ inside the excluded window would overturn the spin-based exclusion for that mass."],"forward_implications":["The three high-spin events push the excluded scalar and vector mass ranges across roughly an order of magnitude centered near $10^{-13}$ eV, so any boson dark-matter model in that window must avoid efficient spin-down of stellar-mass black holes.","The remnant search independently disfavors vector bosons near $3\\times10^{-13}$ eV with no reliance on black-hole population or age assumptions.","Because the two methods cover overlapping yet distinct mass regions, their combination narrows the allowed parameter space more than either alone.","Each additional nearby merger remnant or well-spin-constrained binary will enlarge the union of excluded intervals."],"supporting_citations":[{"why":"Supplies the SuperRad waveform model that predicts strain, timescales, and frequency drift for remnant targets.","marker":"[50]"},{"why":"Provides the SuperRad package used to compute $\\chi_{\\max}$ for the spin-based constraints.","marker":"[106]"},{"why":"Introduces the HMM directed-search method and its sensitivity guidelines.","marker":"[85]"},{"why":"Extends the HMM pipeline and fixes the empirically sensitive search mass range.","marker":"[86]"},{"why":"Prior application of the same search; sets the disfavored-confidence framework, veto stages, and injection averaging.","marker":"[75]"},{"why":"Establishes the spin-based superradiance constraint method, including the prior-driven non-exclusion criterion.","marker":"[58]"},{"why":"Catalogs the O4b events and source properties used to select high-spin binaries and remnants.","marker":"[87]"},{"why":"Provides final parameter estimation for the O4c remnant targets.","marker":"[88]"},{"why":"Waveform model NRSur7dq4 used to estimate remnant masses and spins for the targeted events.","marker":"[96]"},{"why":"Defines effective strain amplitude relating the detector response to injection amplitudes.","marker":"[109]"}],"fun_headline_variants":["GW data disfavor vector bosons from two merger remnants","Spin-down of three BHs restricts ultralight boson masses","Two mergers, three spins: new limits on ultralight bosons","No boson clouds around remnants; spins add exclusions","LIGO-Virgo-KAGRA: no ultralight boson signal in mergers"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The load-bearing premise is that the SuperRad waveform model correctly predicts how a vector boson cloud radiates, including its strain amplitude, growth and depletion times, and frequency drift, so that the quoted detection probability really measures sensitivity; the spin-based constraints additionally rest on the assumption that the binaries' black holes are at least $10^5$ years old.","fun_headline_variants_meta":{"raw":{"variants":["GW data disfavor vector bosons from two merger remnants","Spin-down of three BHs restricts ultralight boson masses","Two mergers, three spins: new limits on ultralight bosons","No boson clouds around remnants; spins add exclusions","LIGO-Virgo-KAGRA: no ultralight boson signal in mergers"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000373,"raw_usage":{"total_tokens":2036,"prompt_tokens":1031,"completion_tokens":1005,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":647,"completion_tokens_details":{"reasoning_tokens":910}},"tokens_in":647,"tokens_out":1005,"duration_ms":11282,"temperature":1.0,"reasoning_tokens":910,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-16T00:33:57.495390+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"A detected long-transient signal from either remnant at the frequency and drift predicted by SuperRad for a vector mass in the disfavored range would refute the no-signal claim, as would an independent analysis that recovers such a signal in the same data. Conversely, recomputing the sensitivity with an alternative waveform model that drops the detection probability below 90% would show that the quoted range is a modeling artifact rather than an established absence.","supporting_citations":[{"cited_title":"Jones, N","cited_arxiv_id":null,"evidence_quote":"Extends the HMM pipeline and fixes the empirically sensitive search mass range."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Prior application of the same search; sets the disfavored-confidence framework, veto stages, and injection averaging."}],"review_version":1}