{"id":"b7aae560-486f-4a54-b7d0-d060e97ed5d6","arxiv_id":"2506.00219","paper_version":2,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":5.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":2,"one_line_summary":"Removing low multipoles from Planck CMB maps and from Lambda CDM simulations removes the homogeneity scale discrepancy; the paper attributes this to the low quadrupole, but its maps labeled ell>=2 retain the quadrupole.","lead":"This paper tests whether the CMB's unexpectedly early homogeneity is caused by the low quadrupole and by foreground galaxies. Removing low multipoles from both data and simulations makes observations match Lambda CDM, while masking nearby galaxies only slightly helps.","discovery_kind":"extension","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The quadrupole attribution is unproven: the analysis never isolates ell=2, since the 'ell>=2' maps retain the quadrupole and the 'ell>=5' cut removes four other modes; a dedicated ell>=3 cut is needed.","rationale":"The reader's weakest assumption identifies exactly the same load-bearing concern: the quadrupole attribution is not established because the multipole cuts do not isolate ell=2. This is the right focus. The paper's most novel and distinctive conclusion is that the homogeneity-scale tension is caused by the low quadrupole; if that cannot be disentangled from other low multipoles, the paper reduces to a demonstration that removing low multipoles (in some combination) improves agreement and that foreground masks do not. That weaker result is still supported by the tables, but it is not the headline claim. The proposed concrete test is minimal and unambiguous: add an ell>=3 map. It uses the same machinery already in the paper, so it is computationally cheap. Because the existing ell>=2 result already shows strong improvement, the comparison between ell>=3 and ell>=5 is what will settle whether the quadrupole is the driver. The chi-square estimator ambiguity noted by the reader is secondary and does not affect the p-value-based attribution argument. Thus the reader's CONDITIONAL verdict is appropriate and no change is needed.","tokens_in":14043,"tokens_out":2718,"duration_ms":26910,"concrete_test":"Regenerate the full analysis using a map that retains only ell>=3 (i.e., zero out the spherical harmonic coefficients a_lm for ell=0,1,2), alongside the existing ell>=2 and ell>=5 maps, using the same 1000 synfast realizations, common mask, and jackknife procedure. Compute theta_W, theta_H, S90, their p-values, and reduced chi-squared for this new map. If the ell>=3 results match the ell>=5 results (p-values above 20%, reduced chi^2 near 1), the quadrupole attribution is confirmed. If they resemble the ell>=2 results, or fall between, then the improvement is driven by monopole/dipole removal or by modes ell=3 and ell=4, and the central claim fails.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central claim, stated in the abstract and Section 4, is that removing the CMB quadrupole (ell=2) specifically resolves the homogeneity-scale discrepancy. But no analysis isolates ell=2. The paper constructs maps by removing 'the first two multipoles' and 'the first five multipoles.' Under the standard harmonic indexing used in Eqs. (2.1)-(2.3), the first two multipoles are ell=0 (monopole) and ell=1 (dipole), so the labeled 'ell>=2' map retains the quadrupole, octopole, and all higher modes. The 'ell>=5' map removes ell=0,1,2,3,4 simultaneously, so it cannot separate the quadrupole from the octopole and hexadecapole. Section 3.3 states that 'the removal of the quadrupole (ell=2) is sufficient' to bring agreement, but this statement is not backed by any map in which only ell=2 is removed. Tables 1-3 show that the 'ell>=2' cut (actually removing monopole and dipole) already raises p-values from 1% to 18-23%, and the additional change from 'ell>=2' to 'ell>=5' is modest. Thus the observed improvement could be driven by monopole/dipole removal, by the combination of ell=0-4, or by some other low-mode combination, rather than by the quadrupole itself. The conclusion that 'the observed low value of the CMB quadrupole contributes notably' is therefore a labeling artifact unless an explicit ell=2-only excision is performed and tested.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper revisits the claimed discrepancy between the observed CMB homogeneity scale and the ΛCDM prediction, using the angular two-point correlation function, the homogeneity index H(θ), and the slope S90 at θ=90°. The analysis uses the Planck SMICA map with the common mask, 1000 synthetic ΛCDM maps, and jackknife resampling. Two modifications are studied: removal of low multipoles (labeled ℓ≥2 and ℓ≥5) from both data and simulations, and application of three foreground masks (L2023, H2023, and a new Neargal mask). The authors report that low-multipole removal substantially improves data-simulation agreement and conclude that the quadrupole anomaly is the main driver of the homogeneity-scale tension, while foreground masking alone does not resolve the discrepancy.","tokens_in":14369,"tokens_out":7842,"duration_ms":76181,"significance":"If the quadrupole attribution were established, this would be a useful result connecting the homogeneity-scale tension to the well-known low-ℓ CMB anomalies, and the negative foreground-mask result would also be informative. The paper has concrete strengths: it uses 1000 synthetic maps, jackknife uncertainties, multiple estimators, and several masks, and it reports p-values and reduced chi-squared values in a transparent tabular form. However, the central claim is currently not established because the maps labeled ℓ≥2 do not remove the quadrupole under the paper's own harmonic convention, so the specific attribution to the quadrupole is unsupported.","major_comments":[{"comment":"The maps labeled ℓ≥2 are obtained by removing 'the first two multipoles', which in the harmonic expansion of Eq. (2.1) are ℓ=0 and ℓ=1; the quadrupole ℓ=2 is therefore retained in the ℓ≥2 maps, while the ℓ≥5 maps remove ℓ=0,1,2,3,4 simultaneously. Since Table 3 shows that the ℓ≥2 cut alone already raises p-values from 1% to 18–23% (for example, θ_H from 1.0% to 23.0%), the improvement cannot be attributed specifically to the quadrupole; it may be driven by monopole/dipole removal. The abstract and Section 4 claim that the quadrupole 'in particular' drives the effect, but this is not supported by the displayed cuts. Please add an analysis with a genuine ℓ≥3 map (removing ℓ=0,1,2) or an ℓ=2-only excision and compare it with the ℓ≥2 and ℓ≥5 cases.","section":"Sec. 3.1, Table 1/3, Figs. 2/5"},{"comment":"The statement that 'the removal of the quadrupole (ℓ=2) is sufficient' is not backed by the displayed analysis: no map isolates ℓ=2. The ℓ≥5 removal includes ℓ=3 and ℓ=4, so the additional improvement from ℓ≥2 to ℓ≥5 could be due to the octopole and hexadecapole instead of the quadrupole. At minimum, a ℓ≥3 map is needed to isolate the quadrupole contribution, and the text should be revised to describe the actual cuts (ℓ<2 and ℓ<5) rather than 'quadrupole removal'.","section":"Sec. 3.3, first paragraph"},{"comment":"The p-values are computed after removing multipoles that were selected because the full-map statistics are anomalous. These are conditional, post-selection probabilities; the paper does not account for the fact that the low multipoles were chosen a posteriori. Consequently, improved p-values after the cuts are expected if the anomaly resides in the removed modes, but they do not by themselves quantify the evidence for the specific claim that the quadrupole amplitude is anomalously low. Please add a direct test of the quadrupole amplitude (for example, comparing observed C_2 with the ΛCDM distribution) or a likelihood comparison that includes the selection, or explicitly frame the results as a decomposition rather than a significance test.","section":"Sec. 3.1, Tables 3 and 5"}],"minor_comments":[{"comment":"The text says the CMB map is downgraded to Nside=4 but then refers to estimating a_lm from ℓ=0 to ℓ=4·(Nside=64); this is contradictory and should be clarified, including the actual maximum multipole used.","section":"Sec. 2.3"},{"comment":"The definition of P(<θ) appears malformed in the typeset equation; the ratio of the observed integrated correlation to the Poisson expectation should be written explicitly with distinct symbols for the numerator and denominator.","section":"Eq. (2.13)"},{"comment":"In Eq. (3.2), Δ_i is already divided by σ_W(i); if the matrix C_{ij} in Eq. (3.1) is the full covariance, the normalization is double-counted. Please specify whether C is the correlation matrix or define Δ_i without the pre-division.","section":"Eqs. (3.1)-(3.2)"},{"comment":"The method used to compute the p-values should be stated explicitly, including whether they are one-sided or two-sided and whether they are the fraction of 1000 simulations with statistics more extreme than the data.","section":"Tables 3, 5, and 8"},{"comment":"The wording 'first two multipoles' and 'first five multipoles' is ambiguous and has led to the quadrupole misidentification; please consistently define the cuts as ℓ<2 and ℓ<5, and fix the repeated typo 'remotion' (e.g., Section 2.3 title and the abstract).","section":"Throughout"}],"recommendation":"major_revision","confidential_remarks":"The central problem appears to be a labeling and interpretation error: the authors treat ℓ≥2 maps as removing the quadrupole, but under their own Eqs. (2.1)-(2.3) those maps retain it. This is fixable with a new ℓ≥3 run and a revised interpretation, so I do not recommend rejection, but the abstract's main claim overstates the evidence until that analysis is provided."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"The paper does something useful: it checks whether the reported CMB homogeneity-scale tension survives low-multipole removal and foreground masking. The p-values for theta_W, theta_H, and S90 under these cuts are new, and the negative result on nearby-galaxy masking is a legitimate data point. The jackknife and 1000-map simulation pipeline is standard and appears competently executed. I buy the broad conclusion that removing the lowest multipoles brings the SMICA map and Lambda CDM into much better agreement. That is worth saying clearly.\n\nThe soft spot is the abstract's and Section 4's claim that the quadrupole specifically is responsible. That claim is not supported by the analysis as presented. The maps labeled ell>=2 actually remove monopole and dipole only; the quadrupole is retained. The ell>=5 maps remove ell=0-4 simultaneously, so they cannot separate the quadrupole from the octopole and hexadecapole. Section 3.3 asserts that removing the quadrupole is sufficient, but there is no map that removes only ell=2. Tables 1-3 show that removing monopole and dipole already lifts p-values to 18-23%, and the additional change from ell>=2 to ell>=5 is modest. So the improvement could come from monopole/dipole removal, from the combined ell=0-4 band, or from something else. The paper needs an explicit ell=2-only (or ell>=3) cut before it can attribute anything to the quadrupole. This is a labeling error in the text, not a fabrication, but it is load-bearing for the advertised result.\n\nThere is also a mild circularity concern: removing modes that are known to be anomalous and then finding improved agreement is partly a restatement of the known low-quadrupole anomaly. That does not sink the paper, but it tempers the novelty. The foreground-masking results are clean and the negative result is honest.\n\nI would send this to a serious referee. The numerical machinery is fine and the low-multipole removal effect is real; the paper just overclaims the quadrupole attribution. A revision that adds an ell>=3 cut or rewrites the conclusions to say 'low multipoles' rather than 'the quadrupole' would fix most of the problem. The chi-square definition is also a bit under-specified, but that is a minor fix.\n\nRecommended verdict: major revision, not rejection. The paper is for people who work on CMB anomalies and homogeneity-scale tests; they will get a useful dataset of p-values even if the interpretive claim needs to be dialed back.","headline":"A careful reanalysis of the CMB homogeneity scale whose cleanest result gets overshadowed by an unproven quadrupole claim; the low-multipole removal works, but the paper never isolates ell=2.","tokens_in":15001,"tokens_out":2006,"would_cite":false,"duration_ms":21893,"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 the mismatch between the CMB homogeneity scale and the ΛCDM model is not an independent anomaly: it is carried by the low observed quadrupole, and masking foregrounds from nearby galaxies does not remove it.","keywords":["cosmic microwave background","homogeneity scale","CMB quadrupole anomaly","angular correlation function","homogeneity index","Lambda-CDM model","extragalactic foreground","Planck SMICA map"],"falsifier":"Construct maps that excise only the quadrupole — remove $\\ell = 2$ while keeping $\\ell = 0,1,3,4,5,\\dots$ — from both the SMICA map and the ΛCDM simulations, and recompute $\\theta_W$, $\\theta_H$, and $S_{90}$. The paper's claim predicts the p-values jump above the ~16% no-tension threshold with only this mode removed; if they stay near 1–3%, the improvement is a collective low-multipole effect and the quadrupole attribution fails. A complementary check is to replace the simulated $\\ell=2$ coefficients with the observed SMICA quadrupole values and verify that the synthetic homogeneity scales move to the observed ones.","tokens_in":13817,"feed_emoji":"🌌","tokens_out":15720,"duration_ms":141138,"temperature":0.7,"pith_summary":"This paper tries to pin down what causes the mismatch between the angular scale at which the cosmic microwave background becomes homogeneous and the prediction of the standard ΛCDM model. Working with Planck SMICA maps, a thousand ΛCDM simulations, and the homogeneity-index estimator $H(\\theta)$, the authors find that cutting the lowest angular modes from both data and simulations erases the discrepancy: the derived scales $\\theta_W$, $\\theta_H$, and the slope $S_{90}$ move into agreement, with p-values rising from about 1% to 12–33%. They attribute this to the anomalously low CMB quadrupole, because the simulations carry substantial quadrupoles whose removal shifts them toward the data. They also test the new extragalactic foreground traced by nearby galaxies and find that masking those regions improves consistency only slightly and cannot resolve the disagreement. If the paper is right, the homogeneity-scale tension is not an independent challenge to ΛCDM but a symptom of the well-known quadrupole anomaly.","feed_headline":"Low CMB quadrupole, not nearby galaxies, drives homogeneity tension","feed_subtitle":"Cutting the lowest multipoles from both maps and simulations removes the mismatch; masking nearby galaxies does not.","key_machinery":"The argument runs through the homogeneity-index chain adapted from earlier work: the two-point correlation function written as $W_\\eta(\\theta) = 1 + C(\\theta)$, the scaled counts-in-spheres $P(<\\theta)$, the homogeneity index $H(\\theta) = 2\\,d\\ln P(<\\theta)/d\\ln\\Omega(<\\theta)$, and the derived quantities — the crossing scale $\\theta_W$ where $W_\\eta=1$, the homogeneity scale $\\theta_H$ where $H=0$, and the slope $S_{90}$ at $\\theta=90^\\circ$. The decisive comparison is between the Planck SMICA map and 1000 synthetic ΛCDM maps, with data variance estimated from 128 (and 256) jackknife sky regions and disagreement quantified by the p-value of each observed estimator in the simulation distribution. Low-multipole removal is implemented by estimating the $a_{\\ell m}$ coefficients from low-resolution maps through a $\\chi^2$ fit with a mode-coupling matrix, then rebuilding maps that keep only $\\ell \\geq 2$ or $\\ell \\geq 5$; the three foreground masks respectively block large spiral galaxies, higher-redshift extensions of that foreground, and the most nearby galaxies.","core_discovery":"The paper's central claim is that the homogeneity-scale discrepancy in the CMB is caused by the low value of the observed quadrupole, not by foreground emission associated with nearby galaxies. On the full SMICA map the homogeneity estimators $\\theta_W$ and $\\theta_H$ sit at p-values near 1% relative to the ΛCDM simulations, a mild tension by the paper's own calibration, while the three foreground masks — L2023, H2023, and a new Neargal mask built from the Nearby Galaxy Catalogue — leave the estimators essentially unchanged, with $\\theta_H$ p-values of about 1–3%. Removing the lowest multipoles from both data and simulations, however, raises the p-values to 12–33%, inside the regime the paper treats as consistent with the model. The direction of the effect carries the argument: the observed quadrupole is already very small, so cutting it barely changes the data, whereas the simulated quadrupoles are substantial and their removal shifts the simulated homogeneity scales toward the observed values. The authors read this as direct evidence that the low homogeneity scale is the quadrupole anomaly appearing in a different observable.","pith_inferences":["A sharper test would isolate $\\ell = 2$: the maps analyzed here either keep the quadrupole in (the $\\ell \\geq 2$ case) or remove it together with the monopole, dipole, octopole, and $\\ell = 4$ modes (the $\\ell \\geq 5$ case), so the specific attribution of the effect to the quadrupole alone is an interpretation, not a measured contrast.","If an $\\ell = 2$-only test confirms the attribution, the homogeneity scale stops being a separate cosmological constraint: any mechanism that explains the low quadrupole — modified initial conditions, a physical foreground, or non-standard geometry — would automatically reconcile the large-angle CMB with ΛCDM.","The same analysis could be run on the WMAP data and on the other Planck component-separation maps; the paper's claim predicts the same jump in p-values across all of them, whereas a foreground origin would predict method-dependent behavior."],"forward_implications":["The homogeneity-scale discrepancy in $\\theta_W$, $\\theta_H$, and $S_{90}$ disappears once the lowest multipoles are removed from both data and simulations, with p-values rising from about 1% to 12–33%.","Masking the new extragalactic foreground traced by nearby galaxies does not resolve the discrepancy; at best it improves consistency marginally, with $\\theta_H$ p-values staying near 3% on the full map.","The improvement comes mainly from the simulations: the data quadrupole is already small, so removing it barely changes the data, whereas cutting the substantial simulated quadrupoles pulls the ΛCDM homogeneity scales down toward the observed values.","With the low multipoles cut, the reduced $\\chi^2$ for both $W_\\eta(\\theta)$ and $H(\\theta)$ drops to roughly 1 for the data against ΛCDM, whether or not the Neargal mask is also applied.","If the paper is right, the homogeneity scale adds no independent tension to ΛCDM beyond the quadrupole anomaly itself."],"supporting_citations":[{"why":"Supplies the homogeneity-scale estimator $H(\\theta)$ and the baseline CMB homogeneity measurement that this paper reproduces and extends.","marker":"[51]"},{"why":"Provides the Planck PR3 ΛCDM best-fit spectrum used to generate the 1000 synthetic maps and sets the p-value conventions for assessing agreement.","marker":"[42]"},{"why":"Defines the new extragalactic foreground around large spiral galaxies and supplies the L2023 mask.","marker":"[45]"},{"why":"Gives the $>3\\sigma$ detection of the foreground and the low-multipole $a_{\\ell m}$ estimation methodology that the paper follows when cutting low multipoles.","marker":"[47]"},{"why":"Supplies the H2023 foreground model and mask used in the masking comparisons.","marker":"[48]"},{"why":"Provides the Nearby Galaxy Catalogue from which the new Neargal mask is built.","marker":"[55]"},{"why":"Defines the $\\chi^2$ statistic used for all consistency numbers.","marker":"[56]"},{"why":"Sets the p-value methodology for judging how unusual the observed homogeneity estimators are under ΛCDM.","marker":"[57]"}],"fun_headline_variants":["Quadrupole, not galaxies, causes CMB homogeneity mismatch","CMB homogeneity gap persists despite galaxy masking","Low quadrupole drives CMB homogeneity discrepancy","Galaxy masking fails; quadrupole removal fixes CMB homogeneity"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The conclusion that the quadrupole in particular drives the tension rests on an untested premise, since Section 3.3 asserts that removing $\\ell=2$ suffices while the maps actually analyzed either keep the quadrupole in (the $\\ell \\geq 2$ case) or remove it together with four other low modes (the $\\ell \\geq 5$ case).","fun_headline_variants_meta":{"raw":{"variants":["Quadrupole, not galaxies, causes CMB homogeneity mismatch","CMB homogeneity gap persists despite galaxy masking","Low quadrupole drives CMB homogeneity discrepancy","Galaxy masking fails; quadrupole removal fixes CMB homogeneity"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000311,"raw_usage":{"total_tokens":1813,"prompt_tokens":1031,"completion_tokens":782,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":647,"completion_tokens_details":{"reasoning_tokens":717}},"tokens_in":647,"tokens_out":782,"duration_ms":9170,"temperature":1.0,"reasoning_tokens":717,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-07T12:10:32.385537+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Construct maps that excise only the quadrupole — remove $\\ell = 2$ while keeping $\\ell = 0,1,3,4,5,\\dots$ — from both the SMICA map and the ΛCDM simulations, and recompute $\\theta_W$, $\\theta_H$, and $S_{90}$. The paper's claim predicts the p-values jump above the ~16% no-tension threshold with only this mode removed; if they stay near 1–3%, the improvement is a collective low-multipole effect and the quadrupole attribution fails. A complementary check is to replace the simulated $\\ell=2$ coefficients with the observed SMICA quadrupole values and verify that the synthetic homogeneity scales move to the observed ones.","supporting_citations":[{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Supplies the homogeneity-scale estimator $H(\\theta)$ and the baseline CMB homogeneity measurement that this paper reproduces and extends."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Provides the Planck PR3 ΛCDM best-fit spectrum used to generate the 1000 synthetic maps and sets the p-value conventions for assessing agreement."},{"cited_title":"E., Boero, E","cited_arxiv_id":null,"evidence_quote":"Defines the new extragalactic foreground around large spiral galaxies and supplies the L2023 mask."},{"cited_title":"K., Lambas, D","cited_arxiv_id":null,"evidence_quote":"Gives the $>3\\sigma$ detection of the foreground and the low-multipole $a_{\\ell m}$ estimation methodology that the paper follows when cutting low multipoles."},{"cited_title":"K., Boero, E","cited_arxiv_id":null,"evidence_quote":"Supplies the H2023 foreground model and mask used in the masking comparisons."},{"cited_title":"D., Makarov, D","cited_arxiv_id":null,"evidence_quote":"Provides the Nearby Galaxy Catalogue from which the new Neargal mask is built."},{"cited_title":"& Hughes, D","cited_arxiv_id":null,"evidence_quote":"Defines the $\\chi^2$ statistic used for all consistency numbers."},{"cited_title":"& Huterer, D","cited_arxiv_id":null,"evidence_quote":"Sets the p-value methodology for judging how unusual the observed homogeneity estimators are under ΛCDM."}],"review_version":1}