{"id":"0326a1f0-8ada-4e6b-90d3-05b9c837a1b0","arxiv_id":"2607.22461","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":5.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":5,"one_line_summary":"A 2.8σ detection of the integrated Sachs-Wolfe effect in the Quaia×Planck cross-correlation gives A_ISW = 1.69 ± 0.61, moderately above ΛCDM and unexplained by current CPL dark-energy models.","lead":"This paper cross-correlates Planck CMB temperature maps with the Quaia quasar catalogue to measure the integrated Sachs-Wolfe effect, detecting it at 2.8σ with an amplitude about 1.7 times the standard cosmological-constant prediction. The measurement is interesting because a slightly high ISW amplitude is a long-standing hint in cosmology, and this analysis tests whether evolving dark-energy models favoured by recent DESI/DES data can explain it — they cannot.","discovery_kind":"new_application","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Detection significance is sensitive to ℓmax: the linear-regime cut (ℓmax=100) yields 2.1σ versus the headline 2.8σ; the high-ℓ outlier may inflate the claim.","rationale":"The reader's weakest assumption is the fixed quasar bias model (Sect. 2.2, Eq. 2). That concern affects the amplitude calibration: substituting the Laurent et al. (2017) bias changes A_ISW from 1.69 to 2.02 but leaves the detection significance essentially unchanged (2.78σ vs 2.78σ), as reported in Sect. 3.3. Thus the bias model does not threaten the central '2.8σ detection' claim. The more load-bearing issue is that the detection significance itself depends on the chosen ℓmax and on the treatment of the ℓ≈200 outlier. The paper's own robustness tests show this dependence, but the abstract presents only the fiducial 2.8σ. A conservative scale cut (ℓmax=100) gives 2.1σ, which is still suggestive but not the quoted 'detection'. Since the central claim is the significance, this scale-sensitivity should be explicitly reported and, if the outlier is not genuinely statistical, the conclusion should be softened. This does not change the overall CONDITIONAL verdict — the paper is a credible measurement but requires revision to honestly represent the significance — so the verdict remains UNCHANGED relative to the reader's assessment.","tokens_in":14353,"tokens_out":11823,"duration_ms":137367,"concrete_test":"Re-fit the total-sample cross-spectrum using the fiducial 6-bin scheme but omitting the single bin centered near ℓ≈200, and separately recompute with ℓmax=100 while keeping the same binning logic (if possible). Compare the resulting A_ISW and S/N from Eq. 10 with the fiducial values. Additionally, quantify the fraction of the 2000 mocks that show a ≥2σ outlier in that specific bin; if the fraction is large, the outlier is plausibly statistical. If the S/N drops below ~2.5σ once the outlier is removed or with the linear-scale cut, the abstract should quote the scale-cut-independent significance rather than 2.8σ.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central claim is the 2.8σ ISW detection. In Sect. 3.1 the fiducial ℓmax=300, 6-bin analysis gives A_ISW=1.69±0.61, i.e. S/N=2.78. In Sect. 3.3, restricting to ℓmax=100 (5 bins) gives A_ISW=1.30±0.61, i.e. S/N=2.13. The drop in S/N is about 0.65σ, yet the abstract quotes the 2.8σ value without qualification. Figure 4 shows the significance varies with ℓmax (2.13, 2.70, 2.77, 2.86 for ℓmax=100,500,767(7 bins),767(6 bins)). The increase at higher ℓmax is driven by multipoles 100–300 where the ISW template is small and where the authors note 'an outlier data point ... at ℓ≈200 even for the total sample' (Sect. 3.3). If that outlier is a systematic — unresolved CIB, tSZ, or nonlinear Rees-Sciama contribution not captured by the template or the mock covariance — the fitted amplitude is biased high and the detection significance is inflated. The paper's statement that results are 'robust' refers to amplitude consistency within 1σ, not to significance. Since the headline is a detection significance, this scale-dependence is the most load-bearing weakness.","agreement_with_reader":"partial"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper presents a tomographic cross-correlation analysis between the Quaia quasar catalogue and Planck 2018 CMB temperature maps, measuring the integrated Sachs-Wolfe (ISW) effect. Using a pseudo-Cℓ pipeline with 2000 mock-based covariances, the authors report a best-fit ISW amplitude A_ISW = 1.69 ± 0.61 (2.78σ) relative to the Planck ΛCDM prediction for the total quasar sample, with low-z and high-z bin amplitudes of 1.19 ± 0.56 and 2.86 ± 1.62, respectively. They also compare against two w0waCDM models favoured by recent DESI/DES data and find these models do not explain the mildly elevated amplitude. Robustness tests include alternative CMB component-separation maps, a second quasar bias model, North/South sky splits, and variations of multipole binning and ℓmax. The central claim is an ISW detection at ~2.8σ with an amplitude consistent with ΛCDM at 1.1σ, and the paper argues that current data cannot distinguish ΛCDM from the CPL models but prefer an amplitude above both.","tokens_in":14732,"tokens_out":3131,"duration_ms":38508,"significance":"If taken at face value, the result provides one of the most extensive tomographic ISW measurements to date, using the Quaia catalogue's large sky area and redshift depth. The analysis is methodologically transparent: it uses the public NaMaster pseudo-Cℓ estimator, a covariance from 2000 correlated Gaussian mocks with Hartlap correction, and external cosmological parameters from Planck and DESI/DES, avoiding circular use of the ISW amplitude. The robustness tests, especially the use of NILC and SEVEM CMB maps and the alternative bias model, strengthen confidence in the amplitude stability. The paper also gives a useful comparison with w0waCDM models motivated by recent dark-energy hints. However, the headline significance is scale-dependent—dropping to 2.1σ when ℓmax is restricted to the linear regime—and the fixed quasar bias normalization adds unquantified systematic uncertainty to the amplitude. These issues must be addressed before the detection claim is fully supported.","major_comments":[{"comment":"The headline detection significance of 2.8σ (abstract, §3.1) is not stable under the choice of ℓmax: restricting to ℓmax=100 with 5 bins gives A_ISW=1.30±0.61, i.e. S/N=2.13σ, while the fiducial ℓmax=300 yields 2.78σ. The paper states in §3.3 that results are 'robust', but this robustness applies to the amplitude, not to the significance. Since the central claim is a detection at 2.8σ, this scale dependence is load-bearing. Please report the linear-regime significance as a primary result, or provide a quantitative justification for including multipoles 100–300 (e.g., validation of the template and covariance on those scales) rather than relying on a 1σ amplitude consistency.","section":"§3.1, §3.3, Fig. 4"},{"comment":"The quasar bias normalization b0=1.26 is fixed from Piccirilli et al. without propagating its uncertainty. The ISW cross-spectrum template is proportional to b(z), so A_ISW scales inversely with b0. The alternative bias model of Laurent et al. shifts A_ISW from 1.69 to 2.02 (§3.3), illustrating the sensitivity. Since the paper uses A_ISW to assess dark-energy models, the error on A_ISW should include the bias uncertainty. Please marginalize over b0 with a prior informed by Piccirilli et al., or at least quote a systematic error from the allowed b0 range. Without this, the reported 1σ error is underestimated.","section":"§2.2, Eq. (2); §3.3"},{"comment":"The increase in significance from ℓmax=100 to ℓmax=300 appears driven by multipoles 100–300, where the ISW template is small relative to the error bars and where the paper notes 'an outlier data point ... at ℓ≈200' even for the total sample. The text dismisses this as a statistical fluctuation, but no explicit test is shown—e.g., removing that bin from the fit, comparing with a map without the outlier, or examining known foreground residuals (CIB, tSZ) on those scales. Given that this outlier contributes to the 2.8σ claim, its nature should be tested rather than assumed, and the resulting significance change should be reported.","section":"§3.3, Fig. 3"}],"minor_comments":[{"comment":"Typo: 'comsic maps' should be 'cosmic maps'.","section":"§4"},{"comment":"The y-axis label 'Custume Normalized dN/dz' appears to be a typo; should likely be 'Custom Normalized dN/dz' or simply 'Normalized dN/dz'.","section":"Fig. 2"},{"comment":"The description of the mask construction is clear, but it would help to state the total unmasked sky fraction for the joint masks used in the fiducial analysis, as this affects cosmic variance and the interpretation of the significance.","section":"§2.5"},{"comment":"The wCDM model with w=−0.65 is mentioned as 'ad-hoc' and likely ruled out by other constraints; consider removing it or explicitly labeling it as an illustrative exercise, since it could be mistaken for a viable scenario.","section":"§3.2"}],"recommendation":"major_revision","confidential_remarks":"The reader's conditional verdict is well supported: the paper is methodologically sound in its use of mocks and robustness tests, but the headline significance is not as robust as the abstract suggests. The two main issues—ℓmax dependence of S/N and the fixed bias normalization—are fixable within the manuscript's scope by rephrasing the claim and adding a systematic error term. The paper would be a solid contribution after these revisions; it is not a reject."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"The useful take: this is the first ISW cross-correlation using the full-sky Quaia quasar catalogue, and it is a solid, transparent pipeline — 2000 correlated mocks, covariance, checks across CMB maps, binning, ℓmax, hemispheres, and two bias models. The fiducial result is A_ISW=1.69±0.61 (2.8σ) for the total sample. That is a new measurement, not a new method, and it is consistent with earlier QSO-based ISW amplitudes from Stölzner and Xia.\n\nNow the soft spots, in order of severity.\n\nThe significance is not as stable as the paper claims. Restricting to ℓmax=100 (5 bins) drops S/N to 2.1σ; the range across ℓmax choices is 2.1–2.9σ. The paper calls this 'robust' because the fitted amplitude shifts by less than 1σ, but the headline is a detection significance, and that is what changes. They note an outlier at ℓ≈200; if that is a systematic — something like unresolved CIB or tSZ that is not in the mock covariance — the central value and S/N are both inflated. They argue it is statistical, and it might be, but the abstract should not quote 2.8σ unqualified.\n\nThe amplitude is hostage to the bias normalization. They fix b0=1.26 with no error. Since A_ISW scales inversely with b(z), an alternative bias model changes A to 2.02±0.73. The S/N stays around 2.8 because the error scales too, but any statement about 'moderately stronger than ΛCDM' depends on a bias model you have not marginalized over.\n\nThe abstract's 'discrepancy' language overstates the body's 1.1σ consistency with ΛCDM. The CPL models from DESI/DES predict lower ISW, and the data prefer higher, but as they state in §4, ISW alone cannot discriminate. So the dark-energy discussion is honest in the text but the abstract spins it.\n\nAlso: no code or data provided for end-to-end checks, so confidence is moderate. The mocks are generated from the same theory spectra, which is standard for covariance.\n\nThe measurement itself is useful as a Quaia ISW data point, and the comparison with w0waCDM is a reasonable advertisement for future surveys. I would send it to a serious referee — the flaws are fixable in revision (quote the ℓmax=100 number in the abstract, propagate b0 uncertainty, tone down 'discrepancy'). It deserves referee time, not desk rejection.","headline":"A careful new ISW measurement from Quaia, but the headline 2.8σ is softer than it looks — the linear-regime cut gives 2.1σ, and the amplitude is sensitive to the bias model; still worth sending to a referee.","tokens_in":15279,"tokens_out":3211,"would_cite":true,"duration_ms":33853,"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":"The ISW effect is detected at 2.8σ in Quaia quasar–Planck CMB cross-correlations, with amplitude 1.69 ± 0.61, and the evolving-dark-energy models preferred by recent BAO and lensing data cannot explain the excess.","keywords":["integrated Sachs-Wolfe effect","ISW","Quaia quasars","CMB cross-correlation","dark energy","w0waCDM","tomography","Planck CMB"],"falsifier":"Measure the actual large-scale bias of the Quaia z≈0.97 bin from its autocorrelation or from cross-correlation with CMB lensing, and compare with 1.26; if the true b0 were, say, 1.5, the central amplitude would fall to about 1.4, within 1σ of ΛCDM. A bias-independent high-redshift ISW measurement using a different tracer with precisely known bias at z≈2 would settle whether the 2.86 ± 1.62 amplitude is physical.","tokens_in":14248,"feed_emoji":"🔭","tokens_out":7762,"duration_ms":86395,"temperature":0.7,"pith_summary":"This paper tries to establish that the integrated Sachs–Wolfe effect—the imprint of late-time gravitational-potential decay on CMB photons—shows up in the cross-correlation between the Quaia quasar catalogue and Planck temperature maps. The central result is a 2.8σ detection with amplitude A_ISW = 1.69 ± 0.61 times the Planck ΛCDM prediction, meaning the measured signal is somewhat stronger than the standard model expects but still statistically consistent with it. The paper also argues this amplitude is stable when the CMB map, quasar bias model, sky region, multipole range, or binning scheme is changed, and that the w0waCDM dark-energy models favoured by recent BAO and weak-lensing data predict a lower signal and therefore do not explain the excess. A reader should care because ISW measurements are one of the few direct fingerprints of dark energy's effect on structure growth, and a true excess would point beyond ΛCDM.","feed_headline":"Quasars and CMB reveal dark-energy signal at 2.8 sigma","feed_subtitle":"The signal is 1.69x the cosmological-constant prediction and holds under changes in maps, bias, and binning.","key_machinery":"The central object is the tomographic angular cross-power spectrum C^{Tg}_ℓ between Planck temperature and Quaia quasar overdensity maps, computed with a standard pseudo-Cℓ spherical-harmonic mode-coupling estimator on a common sky mask and fitted by a single amplitude A_ISW. The theoretical template combines an ISW kernel W_ISW(χ) ∝ H(z) [1 - f(z)]—the rate at which gravitational potentials decay with time—with a quasar kernel b(z) dN/dz, where the quasar bias is assumed to follow b(z) = b0 / D(z) with b0 = 1.26. The covariance is estimated from 2000 correlated Gaussian mock realizations of the CMB and quasar density fields, so the fit accounts for cosmic variance and shot noise. This machi","core_discovery":"The paper's central claim is that the ISW cross-correlation signal is present in the Quaia × Planck data at 2.8σ significance, with best-fit amplitude A_ISW ≈ 1.69 ± 0.61 relative to the fiducial Planck ΛCDM template. Split by redshift, the low-z bin (z̄ ≈ 0.97) gives 1.19 ± 0.56, the high-z bin (z̄ ≈ 2.10) gives 2.86 ± 1.62, and a joint fit over both bins gives 1.38 ± 0.53. The w0waCDM models used in the analysis, with parameters taken from recent baryon-acoustic-oscillation and weak-lensing constraints, predict ISW spectra lower than ΛCDM by at most about 20%, and fitting them to the data does not improve χ²; in fact the data prefer an amplitude above the ΛCDM template, in the opposite dir","pith_inferences":["A direct measurement of Quaia's quasar bias normalization at low redshift would be the fastest check: since the fitted amplitude scales as 1/b0, an upward correction to b0 would pull A_ISW toward 1 and the excess would largely dissolve; a downward correction would strengthen it.","The high-z bin is where the excess is most pronounced (2.86 ± 1.62) and where shot noise is largest; a dedicated wide-area sample at z≈2 with much higher tracer density could separate a genuine high-z ISW excess from a statistical outlier.","If the excess persists after bias and systematics are controlled, it points toward models that produce faster late-time potential decay than ΛCDM (for example higher Ωm or σ8, or w > -1), the opposite direction from the w0waCDM fits favoured by current BAO and lensing data; this would be a tension worth pursuing, though the paper does not draw that conclusion."],"forward_implications":["If the central claim is right, the ISW effect is detected in the largest-volume quasar sample to date, providing an independent, growth-based confirmation that gravitational potentials decay at late times as ΛCDM predicts.","The tomographic split indicates the signal is not carried only by low redshift: the high-z bin (z̄≈2.10) yields A_ISW = 2.86 ± 1.62, consistent with previous quasar-ISW estimates, though with large uncertainty.","The measured amplitude exceeds ΛCDM by about 1.1σ, so the result is not evidence against ΛCDM, but it narrows the room for models, like the w0waCDM ones used here, that predict a weaker ISW signal.","The robustness tests—three CMB maps, an alternative bias model, hemisphere splits, and varying ℓmax and bin count—all keep A_ISW within 1σ of the baseline, supporting the interpretation that the cross-correlation is not dominated by a particular analysis choice.","Future wide-area surveys with higher quasar densities will reduce shot noise and can test whether the excess grows or shrinks, which the paper identifies as the path to distinguishing dark-energy models."],"fun_headline_variants":["ISW effect detected at 2.8 sigma with quasars","Quasar–CMB probe reveals ISW signal 1.69 times ΛCDM","Dark energy mystery: quasar cross-correlation finds 2.8σ signal","Quasars and CMB hint at stronger dark-energy signal"],"cache_read_input_tokens":2304,"weakest_assumption_plain":"The load-bearing assumption is the fixed quasar clustering bias, b0 = 1.26 in the model b(z) = b0 / D(z); if the true bias differs, the fitted signal amplitude shifts by the inverse factor and the dark-energy comparison changes, even though the detection significance does not.","fun_headline_variants_meta":{"raw":{"variants":["ISW effect detected at 2.8 sigma with quasars","Quasar–CMB probe reveals ISW signal 1.69 times ΛCDM","Dark energy mystery: quasar cross-correlation finds 2.8σ signal","Quasars and CMB hint at stronger dark-energy signal"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000252,"raw_usage":{"total_tokens":1503,"prompt_tokens":953,"completion_tokens":550,"prompt_tokens_details":{"cached_tokens":256},"prompt_cache_hit_tokens":256,"prompt_cache_miss_tokens":697,"completion_tokens_details":{"reasoning_tokens":468}},"tokens_in":697,"tokens_out":550,"duration_ms":5816,"temperature":1.0,"reasoning_tokens":468,"cache_read_input_tokens":256,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-01T04:39:52.364404+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Measure the actual large-scale bias of the Quaia z≈0.97 bin from its autocorrelation or from cross-correlation with CMB lensing, and compare with 1.26; if the true b0 were, say, 1.5, the central amplitude would fall to about 1.4, within 1σ of ΛCDM. A bias-independent high-redshift ISW measurement using a different tracer with precisely known bias at z≈2 would settle whether the 2.86 ± 1.62 amplitude is physical.","supporting_citations":[],"review_version":1}