{"id":"f7f52bc0-0d49-4e41-b4f6-32d410332c0a","arxiv_id":"2607.05784","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":5.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":2,"one_line_summary":"First CMB lensing reconstruction from two years of SPT-3G Summer temperature data yields A_comb = 1.015 ± 0.053 over 50 < L < 2000, consistent with Planck 2018 ΛCDM.","lead":"SPT-3G reconstructs the CMB lensing potential from two years of temperature data over ~2640 deg² of the Summer survey and finds a lensing amplitude consistent with Planck ΛCDM. The result is an early pathfinder for combining Summer, Main, and Wide fields into a ~10,000 deg² SPT-3G lensing map.","discovery_kind":"extension","skeptic_critique":{"model":"grok-4.5","headline":"Temperature-only quadratic-estimator residual foreground bias may not be fully absorbed into the quoted A_comb uncertainty of 0.053.","rationale":"The reader’s weakest_assumption correctly isolates residual foregrounds, instrumental systematics and temperature-only bias corrections as the load-bearing premise for the quoted error bar. The present concern simply sharpens that premise to the dominant known contaminant for T-only reconstruction (extragalactic foreground four-points). No internal inconsistency or circularity is evident; the measurement is a standard external-consistency pathfinder. Because the full null-test and foreground-budget sections are not independently re-verified here, the CONDITIONAL verdict remains the appropriate holding position. The concrete simulation test above would settle the issue without requiring new data.","tokens_in":2120,"tokens_out":519,"duration_ms":34901,"concrete_test":"Inject a realistic suite of correlated tSZ+CIB+radio simulations (Websky or Sehgal maps scaled to SPT-3G Summer depth and beam) into the exact end-to-end pipeline used for the three fields; recompute A_comb after the same masking and bias subtraction; if the mean recovered A_L − 1 exceeds 0.5 × 0.053 the quoted uncertainty must be inflated or additional foreground mitigation validated.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central claim is that the joint three-field Summer analysis yields an unbiased A^comb = 1.015 ± 0.053 (50 < L < 2000) relative to Planck 2018 ΛCDM. For a temperature-only quadratic estimator the four-point signal receives additive contributions from extragalactic foregrounds (tSZ, CIB, radio sources) that are both non-Gaussian and spatially correlated with the true lensing potential. The reported error bar is therefore load-bearing only if residual foreground bias after masking, filtering and any template cleaning is demonstrated to be ≪ 0.03. The early-results framing and the temperature-only nature of the reconstruction leave this control as the least secure premise: if the residual bias reaches even half the statistical uncertainty, both the consistency statement and the pathfinder claim for a future ~10 000 deg² combination are weakened. Mean-field, N0 and N1 corrections address Gaussian and connected CMB terms but do not automatically cancel the foreground four-point; that cancellation must be shown separately.","agreement_with_reader":"agree"},"referee_report":{"model":"grok-4.5","summary":"This paper presents the first CMB lensing potential reconstruction from the SPT-3G Summer survey using two years of temperature-only data over ~2640 deg² (three fields of 1210, 570, and 860 deg²). A joint analysis yields a lensing amplitude A^comb = 1.015 ± 0.053 relative to a fiducial Planck 2018 ΛCDM cosmology for 50 < L < 2000. The work uses a standard quadratic estimator with N0/N1 bias corrections and mean-field subtraction, and frames the result as a pathfinder for combining Summer with Main and Wide fields toward a ~10 000 deg² SPT-3G lensing map.","tokens_in":2229,"tokens_out":984,"duration_ms":17753,"significance":"If the result holds, it is a solid incremental contribution: the first temperature-only lensing reconstruction from the SPT-3G Summer fields, demonstrating consistency with Planck at the ~5% level and quantifying the S/N gain expected from the full ~10 000 deg² combination. The analysis rests on established pipeline elements (quadratic estimator, bias corrections, multi-field combination) and reports a clear, falsifiable amplitude measurement. Strengths include a well-defined multipole range, explicit field combination, and an early-results framing that sets expectations for future polarization-inclusive and multi-survey combinations. The scientific impact is moderate but real for the SPT-3G program and for cross-checks of ΛCDM lensing amplitude.","major_comments":[{"comment":"The central claim A^comb = 1.015 ± 0.053 is load-bearing only if residual extragalactic foreground bias (tSZ, CIB, radio sources) after masking/filtering is demonstrated to be ≪ 0.03. Temperature-only quadratic estimators receive additive non-Gaussian four-point contributions that are spatially correlated with the true lensing potential; N0, N1, and mean-field corrections address Gaussian/connected CMB terms but do not automatically cancel the foreground four-point. The manuscript must quantify residual bias (e.g., via simulations with realistic foregrounds or template-cleaning null tests) and show it is absorbed into or negligible compared with the quoted 0.053 uncertainty. Without that demonstration the consistency statement and the pathfinder claim for the ~10 000 deg² combination are weakened.","section":null},{"comment":"Null-test and systematic-control suite (field-difference, curl-mode, split-half, and foreground-cleaned vs. uncleaned comparisons) must be reported with quantitative pass/fail criteria tied to the 0.053 error bar. The early-results framing leaves these validations as the least secure premise; if any residual systematic reaches even half the statistical uncertainty, both the amplitude and its interpretation relative to Planck 2018 ΛCDM require revision.","section":null},{"comment":"The multipole range 50 < L < 2000 and the field-combination weights are free analysis choices that enter the reported A^comb. The paper should show that the amplitude is stable under reasonable variations of L_min, L_max, and weighting scheme (or fold the variation into the systematic error budget) so that the quoted uncertainty is not understated by analysis choices.","section":null}],"minor_comments":[{"comment":"Clarify the precise definition of the fiducial Planck 2018 ΛCDM spectrum used for A_L (parameter set, lensing spectrum generation code, and any smoothing or binning) so that the ratio is fully reproducible.","section":null},{"comment":"State explicitly whether the three Summer fields are treated as independent or whether inter-field covariance is included in the joint likelihood; a short equation or covariance-matrix description would help.","section":null},{"comment":"Figure captions and axis labels for the reconstructed lensing power spectrum and A_L bandpowers should include the exact multipole binning and the units of the plotted quantities.","section":null},{"comment":"A brief comparison table or paragraph placing the Summer A^comb next to prior SPT-3G Main/Wide and Planck lensing amplitudes would improve context without lengthening the text substantially.","section":null},{"comment":"Typographical consistency: ensure A^comb vs. A_L notation is uniform throughout the abstract, body, and tables.","section":null}],"recommendation":"major_revision","confidential_remarks":"The reader's stress-test concern about residual temperature-only foreground bias is well-founded and is the primary reason for major_revision rather than minor_revision. The paper is otherwise a standard, competent early-results SPT-3G analysis; once the foreground and null-test sections are strengthened to the level expected for a 5% A_L claim, it should be publishable. Scope fit for an astro-ph.CO journal is good. No novelty or citation-pattern concerns."},"author_rebuttal":{"model":"grok-4.5","summary":"We thank the referee for a careful reading and for identifying three concrete issues that affect the robustness of A^comb. We agree that residual extragalactic foreground bias, a quantitative null-test suite, and stability under multipole/weighting choices must be demonstrated at a level commensurate with the 0.053 statistical uncertainty. We will revise the manuscript accordingly: add a residual-foreground bias estimate (simulations and/or template-cleaning tests), report the full null-test suite with pass/fail criteria tied to the error bar, and show stability of A^comb under variations of L_min, L_max, and field weights (or enlarge the systematic budget). These additions strengthen rather than alter the central consistency result and the pathfinder framing for the ~10 000 deg² combination.","responses":[{"response":"We agree. N0/N1 and mean-field subtraction do not cancel the connected foreground four-point, so residual tSZ/CIB/radio bias after masking and filtering must be quantified relative to the 0.053 error. In the revision we will (i) estimate residual bias with end-to-end simulations that include realistic tSZ, CIB, and radio populations at the SPT-3G frequencies and depths, processed through the same mask, filter, and quadratic estimator; and (ii) where feasible, report template-cleaning or frequency-difference null tests that bound the same bias. We will state the residual bias (or upper limit) explicitly and either show it is ≪ 0.03 or fold a residual systematic into the total uncertainty on A^comb. The consistency statement and pathfinder claim will be phrased only after that demonstration is in place. We do not claim the present draft already contains this quantification at the required level.","revision_made":"yes","referee_comment":"The central claim A^comb = 1.015 ± 0.053 is load-bearing only if residual extragalactic foreground bias (tSZ, CIB, radio sources) after masking/filtering is demonstrated to be ≪ 0.03. Temperature-only quadratic estimators receive additive non-Gaussian four-point contributions that are spatially correlated with the true lensing potential; N0, N1, and mean-field corrections address Gaussian/connected CMB terms but do not automatically cancel the foreground four-point. The manuscript must quantify residual bias (e.g., via simulations with realistic foregrounds or template-cleaning null tests) and show it is absorbed into or negligible compared with the quoted 0.053 uncertainty. Without that demonstration the consistency statement and the pathfinder claim for the ~10 000 deg² combination are weakened."},{"response":"We agree that the early-results framing makes a quantitative null-test suite essential. The revision will report field-difference, curl-mode, split-half (e.g., year or detector-set), and foreground-cleaned vs. uncleaned comparisons, each with a stated pass/fail criterion scaled to the 0.053 statistical uncertainty (e.g., residual amplitude consistent with zero within a fraction of that error, or χ²/dof consistent with the null). Any test that fails or that indicates a residual at a non-negligible fraction of the statistical error will be used either to correct the pipeline or to enlarge the systematic error budget on A^comb, with the interpretation relative to Planck 2018 ΛCDM updated accordingly. We will not leave these validations as implicit.","revision_made":"yes","referee_comment":"Null-test and systematic-control suite (field-difference, curl-mode, split-half, and foreground-cleaned vs. uncleaned comparisons) must be reported with quantitative pass/fail criteria tied to the 0.053 error bar. The early-results framing leaves these validations as the least secure premise; if any residual systematic reaches even half the statistical uncertainty, both the amplitude and its interpretation relative to Planck 2018 ΛCDM require revision."},{"response":"We agree that L_min, L_max, and the field-combination weights are analysis choices that can shift A^comb. The revision will present A^comb (or bandpowers) under reasonable variations of L_min, L_max, and weighting (e.g., inverse-variance vs. equal-area or noise-weighted alternatives). If the scatter is small compared with 0.053 we will state that stability explicitly; if not, we will fold the variation into the systematic error budget so that the final uncertainty is not understated by those choices. The baseline 50 < L < 2000 range and the joint-analysis weights will remain the primary result, with the stability tests documented alongside.","revision_made":"yes","referee_comment":"The multipole range 50 < L < 2000 and the field-combination weights are free analysis choices that enter the reported A^comb. The paper should show that the amplitude is stable under reasonable variations of L_min, L_max, and weighting scheme (or fold the variation into the systematic error budget) so that the quoted uncertainty is not understated by analysis choices."}],"tokens_in":1830,"tokens_out":1112,"duration_ms":23267,"standing_objections":[]},"desk_editor":{"model":"grok-4.5","letter":"The punchline is simple: this is the first CMB lensing reconstruction from the SPT-3G Summer survey (two years of temperature, ~2640 deg² in three fields), and the joint analysis gives A_comb = 1.015 ± 0.053 relative to Planck 2018 ΛCDM for 50 < L < 2000. Consistent with one, and framed as a pathfinder for the eventual ~10,000 deg² combination with Main and Wide.\n\nWhat is actually new is the dataset and the survey region, not the technique. Quadratic estimator, multipole cuts, N0/N1/mean-field corrections — all standard. That is fine. A clean early amplitude from a new intermediate-area temperature map is useful pathfinding, and the three-field combination plus the explicit full-survey roadmap are the real contribution. They do not overclaim cosmology impact (no S8 resolution, no new method). Credit earned: a clear, competitive early A_L with low circularity, since the fiducial is external Planck ΛCDM.\n\nThe soft spot is the usual one for temperature-only lensing, and it is real but proportionate. Residual extragalactic four-point (tSZ, CIB, radio) after masking and filtering is not automatically cancelled by mean-field/N0/N1. The quoted 0.053 only holds if residual bias is shown to be well below the statistical error. That is the load-bearing check for this class of papers, not a novel or fatal flaw. If the null tests and foreground budget in the full text support it — as SPT-3G analyses usually do — the consistency statement stands. If those sections are thin, the pathfinder claim softens. I would not treat the stress-test as a show-stopper without seeing those sections fail.\n\nThis is for people working CMB lensing, structure growth, and SPT/ACT/Planck cross-checks. Not a must-read outside that circle. It deserves a serious referee: early survey product, clear measurement, clear path to higher S/N. Send it out.","headline":"First SPT-3G Summer temperature lensing: A_L = 1.015 ± 0.053 vs Planck; solid pathfinder with standard methods, foreground residual is the check that matters.","tokens_in":3535,"tokens_out":535,"would_cite":false,"duration_ms":24748,"reading_group":"maybe","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":[],"pacs":[],"model":"grok-4.5","headline":"SPT-3G Summer temperature data yield the first CMB lensing reconstruction over 2640 deg², with amplitude 1.015 ± 0.053 matching Planck ΛCDM.","keywords":["CMB lensing","SPT-3G","cosmic microwave background","gravitational lensing","quadratic estimator","temperature maps","ΛCDM","South Pole Telescope"],"falsifier":"A null-test residual (curl-mode power, cross-correlation with a foreground tracer, or difference of auto- and cross-spectra) that shifts the recovered A_L by more than the reported 0.053 error bar, or a statistically significant tension with independent Planck or ACT lensing measurements on the same multipole range.","tokens_in":2904,"feed_emoji":"🌌","tokens_out":951,"duration_ms":29226,"temperature":0.7,"pith_summary":"This paper presents the first reconstruction of the cosmic microwave background lensing potential from the SPT-3G Summer survey, using two years of temperature data across three fields that together cover about 2640 square degrees. A joint analysis of those fields measures the lensing amplitude relative to a Planck 2018 ΛCDM fiducial model as A_comb = 1.015 ± 0.053 for multipoles 50 < L < 2000. The result is statistically consistent with the standard cosmological model and is obtained from temperature-only maps after standard quadratic-estimator bias corrections. A sympathetic reader cares because the measurement already reaches roughly 5 percent precision on a large new sky area and explicitly demonstrates that combining the Summer fields with the SPT-3G Main and Wide fields will raise the signal-to-noise for a total survey footprint of roughly 10 000 square degrees. It therefore supplies an early, concrete step toward higher-precision tests of structure growth via CMB lensing.","feed_headline":"SPT-3G Summer maps measure CMB lensing to 5% precision","feed_subtitle":"Joint analysis of 2640 deg² temperature fields matches Planck and previews the full 10 000 deg² survey","key_machinery":"The temperature-only quadratic estimator of the CMB lensing potential, applied field-by-field after N0, N1 and mean-field bias corrections and then combined into a joint amplitude fit over 50 < L < 2000.","core_discovery":"A joint analysis of the three SPT-3G Summer fields yields a CMB lensing amplitude A_comb = 1.015 ± 0.053 relative to a fiducial Planck 2018 ΛCDM cosmology for the multipole range 50 < L < 2000, obtained from the first temperature-only lensing reconstruction of the approximately 2640 deg² Summer survey.","pith_inferences":["The demonstrated control of temperature-only biases over 2640 deg² implies that residual systematics will not dominate once polarization is added, allowing the full SPT-3G program to push A_L precision well below 3 percent.","A 5 percent A_L measurement already begins to limit simple models of modified gravity or massive neutrinos that rescale the lensing amplitude by more than a few percent.","Extending the multipole range above L = 2000 with the same temperature maps, or with future polarization, would test whether the agreement with ΛCDM persists on smaller scales where baryonic feedback becomes relevant."],"forward_implications":["The measured amplitude confirms that large-scale structure growth on the scales probed by 50 < L < 2000 is consistent with Planck 2018 ΛCDM at the 5 percent level.","Adding the Summer fields to the SPT-3G Main and Wide surveys will raise the total lensing signal-to-noise for a combined footprint of roughly 10 000 deg².","Temperature-only reconstructions from SPT-3G alone can already deliver competitive A_L constraints before polarization data are included.","The same maps and pipeline enable forthcoming cross-correlations with external large-scale-structure tracers over the Summer footprint."],"fun_headline_variants":["SPT-3G Summer temperature data reconstruct CMB lensing to 5%","CMB lensing from 2640 deg² SPT-3G Summer fields matches Planck","First SPT-3G Summer CMB lensing reconstruction yields A=1.015±0.053","Two-year SPT-3G Summer temperature data measure CMB lensing to 5%","SPT-3G joint Summer analysis finds lensing amplitude 1.015±0.053"],"cache_read_input_tokens":2304,"weakest_assumption_plain":"Residual foregrounds, instrumental systematics and the temperature-only quadratic-estimator bias corrections are controlled well enough that the quoted uncertainty on the lensing amplitude is not systematically biased.","fun_headline_variants_meta":{"raw":{"variants":["SPT-3G Summer temperature data reconstruct CMB lensing to 5%","CMB lensing from 2640 deg² SPT-3G Summer fields matches Planck","First SPT-3G Summer CMB lensing reconstruction yields A=1.015±0.053","Two-year SPT-3G Summer temperature data measure CMB lensing to 5%","SPT-3G joint Summer analysis finds lensing amplitude 1.015±0.053"]},"model":"grok-4.5","cost_usd":0.026486,"raw_usage":{"total_tokens":4933,"prompt_tokens":730,"num_sources_used":0,"completion_tokens":101,"cost_in_usd_ticks":264860000,"prompt_tokens_details":{"text_tokens":730,"audio_tokens":0,"image_tokens":0,"cached_tokens":128},"completion_tokens_details":{"audio_tokens":0,"reasoning_tokens":4102,"accepted_prediction_tokens":0,"rejected_prediction_tokens":0}},"tokens_in":730,"tokens_out":101,"duration_ms":42100,"temperature":1.0,"reasoning_tokens":4102,"cache_read_input_tokens":128,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-07-08T19:51:29.414453+00:00","model_set":{"reader":"grok-4.5"},"falsifier":"A null-test residual (curl-mode power, cross-correlation with a foreground tracer, or difference of auto- and cross-spectra) that shifts the recovered A_L by more than the reported 0.053 error bar, or a statistically significant tension with independent Planck or ACT lensing measurements on the same multipole range.","supporting_citations":[],"review_version":1}