{"id":"613389db-8111-4996-9d49-b8e1c6f6203a","arxiv_id":"2412.20259","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":2.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":1,"one_line_summary":"Top-quark pole mass values extracted from NNLO fits to ttbar cross-section data are consistent across four PDF sets and with the PDG 2024 average.","lead":"This conference writeup summarizes an NNLO QCD measurement of the top quark's pole mass using LHC data on top-antitop production. The extracted mass values agree with each other across different PDF sets and with the 2024 Particle Data Group average.","discovery_kind":"review","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Neglected qT-subtraction power corrections may bias the extracted top-quark mass beyond the quoted precision; the 1% uncorrelated theory uncertainty does not test the correlated component.","rationale":"The reader's weakest-assumption analysis identifies the same load-bearing concern: the smallness of neglected qT-subtraction power corrections in the MATRIX predictions. The paper's own wording acknowledges the issue and offers Fig. 1 plus a 1% uncorrelated theory uncertainty as mitigation. However, the central claim of compatibility with PDG 2024 and across PDF sets depends on the extracted mass not being biased by these neglected terms. Because the quoted data uncertainty (0.23 GeV) is comparable to the plausible size of a correlated power-correction bias, and because the validation is not comprehensive, this is the most consequential assumption. The concrete test of refitting with STRIPPER predictions would directly settle whether the bias is within the quoted uncertainty. The paper is a proceedings summary, and the full analysis in Ref. [1] is a published JHEP article with public grids, which supports the methodology, but the proceedings alone does not close the gap. The conditional verdict is appropriate; no change is needed.","tokens_in":5241,"tokens_out":7851,"duration_ms":79366,"concrete_test":"Re-run the NNPDF4.0 global fit (or at least the Run 2 differential fit) replacing the MATRIX-based theory predictions with STRIPPER predictions for the same datasets, using the same covariance treatment, and compare the extracted pole mass. If the shift exceeds 0.2 GeV, the neglected power corrections are not negligible at the claimed precision. Alternatively, take the bin-by-bin MATRIX/STRIPPER ratios from Fig. 1 (and, if available, from the same comparison at m_t around 172 GeV) and apply them as a correlated fractional shift to the theory predictions, then refit; if the central value moves by more than 0.1-0.2 GeV, the 1% uncorrelated uncertainty is insufficient to cover the bias.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central extraction relies on MATRIX, which uses qT subtraction and therefore neglects a class of power corrections. The paper states that these are not large based on the comparison to STRIPPER in Fig. 1, but that comparison is limited to a single top-quark mass value (173.3 GeV), a single PDF set (NNPDF3.0), a single scale choice, and two distributions. The actual fit uses multiple datasets, four different PDF+alpha_s sets, and mass values near 172 GeV. The paper adds a 1% uncorrelated theory uncertainty to the covariance matrix, but this only inflates the error bars; it does not correct for a correlated shift in the theory predictions. qT-subtraction power corrections are expected to be smooth functions of M(tt) and y, so any residual difference from STRIPPER acts as a correlated shape distortion that can shift the fitted m_t by an amount not reflected in the quoted central value. With a quoted data uncertainty of 0.23 GeV, a correlated theory bias of a few tenths of a percent could move the extracted mass by 0.2-0.3 GeV, which would affect the compatibility claim. No quantitative bound on the neglected power corrections is provided; the 1% uncertainty is an order-of-magnitude estimate, not a rigorous limit.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"This proceedings contribution summarizes the NNLO QCD extraction of the top-quark pole mass presented in the companion paper Ref. [1]. The authors compute t-tbar cross sections with MATRIX (qT subtraction) and PineAPPL grids, fit total and normalized differential data from Tevatron and LHC, and repeat the fit with four PDF+alpha_s sets: ABMP16, CT18, MSHT20, and NNPDF4.0. The quoted global results include m_t^pole = 172.15 +/- 0.23 (data) +/- 0.08 (PDF) +0.19/-0.49 (scale) GeV for NNPDF4.0, with similar values for the other PDF sets, and the authors conclude that the extracted masses are mutually compatible and compatible with the PDG 2024 value.","tokens_in":5452,"tokens_out":5643,"duration_ms":62750,"significance":"If the result is correct, the paper demonstrates that normalized t-tbar cross sections at NNLO QCD provide a robust extraction of the top-quark pole mass with reduced sensitivity to PDF and alpha_s choices, and the public release of the MATRIX+PineAPPL grids is a concrete reproducibility asset. The comparison against STRIPPER is a useful cross-check, and the statistical procedure is standard rather than circular. The main significance risk is that the quoted 0.23 GeV data uncertainty may be smaller than the unquantified correlated theory bias from qT-subtraction power corrections, so the compatibility claim is not yet established at the stated precision.","major_comments":[{"comment":"The treatment of qT-subtraction power corrections is load-bearing for the central claim. The STRIPPER comparison is performed at a single top-quark mass (173.3 GeV), a single PDF set (NNPDF3.0), a single scale choice, and for only two distributions, whereas the actual fit combines many datasets, four PDF+alpha_s sets, and mass values near 172 GeV. The 1% uncorrelated uncertainty added to the covariance matrix inflates the diagonal errors but does not account for a correlated shift of the theory prediction; since power corrections are expected to be smooth functions of M(tt) and y, such a correlated shift would move the best-fit m_t without being flagged by the chi-square. Please provide a quantitative bound on the correlated component, for example by refitting with r0 = 0.0005 versus r0 = 0.0015 or by comparing the complete set of fitted observables to local-subtraction predictions, and state the resulting shift in m_t.","section":"Sec. 2 / Fig. 1"},{"comment":"The 1% theory uncertainty is described as covering missing power corrections, Monte Carlo numerical integration error, and PineAPPL interpolation-grid error. These three sources have different correlation structures and different scalings: the integration and interpolation errors can in principle be reduced or validated directly, while the power corrections are a systematic effect. Combining them into a single uncorrelated 1% per bin obscures the dominant source and cannot constrain a correlated bias. Please separate these components, quantify each one, and explicitly identify the largest contribution to the quoted theory uncertainty.","section":"Sec. 2, covariance matrix"},{"comment":"The paper states that threshold resummation and Coulomb-gluon effects are expected to be not particularly relevant because the experimental bins are wide, but no numerical test is provided. If any included bin receives non-negligible threshold corrections, the extracted mass could be biased at a level comparable to the quoted 0.23 GeV data uncertainty. Please support this assumption with a quantitative estimate, for example by comparing the effect of excluding the threshold-adjacent bins or by assigning a dedicated systematic uncertainty.","section":"Sec. 2, threshold effects"}],"minor_comments":[{"comment":"The manuscript contains typographical and spacing errors (for example 'Asecondmainingredientofthefitaresuitableexperimentaldata' and 'succesfully'); a careful proofreading pass is needed.","section":"Throughout"},{"comment":"The sentence 'We do not include neither the effects ... nor the effect ...' contains a double negative and should be rephrased.","section":"Sec. 2"},{"comment":"The total uncertainty shown in the inset is not explicitly defined; please state whether it is the quadrature sum of the data, PDF, and scale components.","section":"Sec. 3 / Fig. 2"},{"comment":"The claim that normalized cross sections strongly reduce alpha_s sensitivity is presented qualitatively; one numerical example quantifying the cancellation would make the argument easier to assess.","section":"Sec. 3"},{"comment":"No chi-square per dataset group or global goodness-of-fit value is reported in this manuscript; since Ref. [1] contains the full analysis, a sentence explicitly referring the reader to those numbers would help the standalone readability of this summary.","section":"Sec. 3"}],"recommendation":"major_revision","confidential_remarks":"The manuscript is a conference-proceedings summary, so a short revision with additional quantitative estimates would be sufficient. The main technical concern is inherited from the companion JHEP paper, but it is stated directly in this manuscript and should be addressed here because the quoted precision is central to the compatibility claim. There is no circularity issue; the self-citation of Ref. [1] is a normal genre feature."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Colleague,\n\nThis is a proceedings write-up of the authors' JHEP paper on extracting the top-quark pole mass from NNLO QCD calculations with MATRIX. The main takeaway: the global fit to normalized ttbar differential cross sections gives m_t = 172.15 ± 0.23 (data) ± 0.08 (PDF) +0.19/−0.49 (scale) GeV for NNPDF4.0, with consistent values for MSHT20, CT18, ABMP16, all agreeing with PDG 2024. The paper itself adds no new numerical results beyond Ref [1]; it's a compact summary with useful figures.\n\nWhat it does well: the methodology is clearly described, the choice of normalized cross sections to suppress PDF/alpha_s sensitivity is sensible, and the uncertainty decomposition (data, PDF, scale via a seven-point envelope) is standard. The release of public interpolation grids is a real plus, though no link or checksum is given.\n\nThe soft spot is the treatment of qT-subtraction power corrections. The paper relies on a comparison to STRIPPER (CHM) at a single mass, a single PDF, and two distributions, then adds a 1% uncorrelated theory uncertainty. That does not exclude a correlated shape distortion that could shift m_t by a few tenths of a GeV. This is a legitimate concern, and the paper is honest about it, but it does not provide a quantitative bound. For a proceedings summary this is acceptable—the full analysis is in the JHEP paper—but a referee would want to see that discussion there. Also, the claim that the grids are 'publicly released' is not backed by an actual location.\n\nWho is this for? Someone wanting a quick overview of the state of top-mass extraction from ttbar cross sections, or a starting point for re-fitting with the grids. It is not a standalone research contribution.\n\nRecommendation: as a proceedings contribution it is fine and deserves to appear; the underlying work is important and the summary is honest. If this came to me as a regular journal paper, I would desk reject because the substance is in Ref [1]. For a proceedings, a light review checking consistency with the JHEP paper is enough.","headline":"Useful proceedings summary of a solid NNLO top-mass extraction; no new numbers, but the cross-PDF comparison and released grids make it a handy reference.","tokens_in":6018,"tokens_out":3622,"would_cite":true,"duration_ms":34753,"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 paper reports that NNLO QCD fits to normalized top-pair cross sections yield a top-quark pole mass of $172.15\\pm0.23$ GeV with the NNPDF4.0 set, and compatible values with three other PDF sets and with PDG 2024.","keywords":["top-quark pole mass","NNLO QCD","top-antitop hadroproduction","parton distribution functions","normalized cross sections","qT subtraction","LHC","PDF sets"],"falsifier":"Take the exact set of normalized $M(t\\bar t)$ and double-differential bins used in the fit and compare MATRIX with an independent local-subtraction NNLO code such as STRIPPER at the same scales and mass values; if the bin-by-bin differences, when propagated through the $\\chi^2$ parabola fit, shift $m_t^{\\rm pole}$ by more than the quoted data uncertainty of roughly 0.2-0.3 GeV, then the power-correction assumption is falsified.","tokens_in":5029,"feed_emoji":"⚛️","tokens_out":10749,"duration_ms":96832,"temperature":0.7,"pith_summary":"NNLO QCD predictions for top-antitop pair production, based on the $q_T$-subtraction code MATRIX and interpolated with PineAPPL, are fitted to Tevatron and LHC total and normalized differential cross-section data. The paper claims that the extracted top-quark pole mass is stable across four modern PDF+$\\alpha_s(M_Z)$ sets and is compatible with the PDG 2024 value; for the global fit with NNPDF4.0 the result is $m_t^{\\rm pole}=172.15\\pm0.23\\,(\\mathrm{data})\\pm0.08\\,(\\mathrm{PDF})\\,^{+0.19}_{-0.49}\\,(\\mathrm{scale})$ GeV. This matters because a parameter as basic as the top-quark mass should be reproducible from different parton inputs, and the quoted accuracy is comparable to the current world average. The use of normalized cross sections is central: it cancels much of the $\\alpha_s$ and PDF dependence that otherwise corrupts total-cross-section fits.","feed_headline":"NNLO fit puts top-quark pole mass at 172.15 GeV","feed_subtitle":"Fits to normalized top-pair cross sections agree across four PDF sets and with the PDG 2024 value.","key_machinery":"The load-bearing machinery is a least-squares fit whose theory input is produced by MATRIX, an NNLO QCD code based on the $q_T$-subtraction infrared-singularity method, interfaced to PineAPPL so that many PDF and scale choices can be evaluated from precomputed grids. The data selection restricts to normalized cross sections, mainly as functions of the top-pair invariant mass $M(t\\bar t)$ and rapidity, because normalization suppresses the PDF and $\\alpha_s$ sensitivity. The fit is run at fixed pole-mass values, the three lowest $\\chi^2$ points are interpolated by a parabola, and the minimum defines the extracted mass, with data, PDF, and scale uncertainties separated by repeating the fit across PDF sets and a seven-point scale variation.","core_discovery":"The paper's central claim is that NNLO QCD fits to inclusive and normalized (multi-)differential $t\\bar t+X$ cross sections can determine the top-quark pole mass in the on-shell scheme with an accuracy competitive with the world average, and that the result does not depend on which modern PDF set is chosen as input. The global fit with NNPDF4.0 gives $m_t^{\\rm pole}=172.15\\pm0.23$ GeV from data, with separate PDF and scale uncertainties of $0.08$ GeV and $^{+0.19}_{-0.49}$ GeV; the corresponding values for MSHT20, CT18, and ABMP16 are 171.79, 171.59, and 171.54 GeV, all consistent with each other and with the PDG 2024 value of $172.4\\pm0.7$ GeV. The fits also show that Run 2 differential datasets dominate the constraint, that total cross-section data alone produce a mass value strongly correlated with $\\alpha_s$, and that normalized cross sections suppress that correlation.","pith_inferences":["Beyond the paper: the agreement across PDF sets is a consistency test, not a proof that the 1% per-bin theory uncertainty is correct; a dedicated propagation of the MATRIX-versus-STRIPPER differences through the fit would quantify how much of the mass shift is absorbed by the parabolic interpolation.","Beyond the paper: as HL-LHC data shrink the statistical error, the assumed 1% uncorrelated theory uncertainty will likely become the dominant systematic, motivating a rigorous treatment of $q_T$-subtraction power corrections and possibly resummation near threshold.","Beyond the paper: the same normalized-cross-section strategy could be applied to double-differential distributions in other mass-sensitive variables, or to single-top production, where the PDF and scale cancellation may behave differently."],"forward_implications":["If the claim holds, top-quark pole mass extraction from hadron-collider data no longer depends on which global PDF fit is used, provided normalized differential cross sections are the fitted observables.","Run 2 differential data will continue to dominate the statistical power, so future precision will come from more differential measurements rather than from more inclusive total cross-section points.","Because normalized cross sections suppress the $\\alpha_s$ correlation, the extracted mass can be compared directly with the PDG average without needing a simultaneous $\\alpha_s$ determination.","Total cross-section fits alone remain unreliable for the pole mass unless PDF and $\\alpha_s$ are fitted together, since the mass and coupling are strongly degenerate in that observable."],"supporting_citations":[{"why":"The full JHEP study underlying this proceedings, from which the methodology and numerical results are taken.","marker":"[1]"},{"why":"Provides the MATRIX code that generates the NNLO QCD cross sections used in the fit.","marker":"[2]"},{"why":"Provides the PineAPPL interpolation grids used to evaluate many PDF and scale choices efficiently.","marker":"[3]"},{"why":"Establishes the $q_T$-resummation formalism on which the $q_T$-subtraction method in MATRIX relies.","marker":"[5,6]"},{"why":"Supplies the STRIPPER predictions used to check that the neglected $q_T$-subtraction power corrections are not large for the fitted distributions.","marker":"[10]"},{"why":"One of the four input PDF+$\\alpha_s$ sets (ABMP16) used to test the stability of the extracted mass.","marker":"[13]"},{"why":"One of the four input PDF+$\\alpha_s$ sets (CT18) used to test the stability of the extracted mass.","marker":"[14]"},{"why":"One of the four input PDF+$\\alpha_s$ sets (MSHT20) used to test the stability of the extracted mass.","marker":"[15]"},{"why":"One of the four input PDF+$\\alpha_s$ sets (NNPDF4.0) used for the quoted global-fit central value.","marker":"[16]"},{"why":"The PDG 2024 review providing the preferred top-quark mass value that the fit results are compared with.","marker":"[17]"}],"fun_headline_variants":["NNLO fit pins top mass to 0.23 GeV precision","Top mass from NNLO QCD: PDF set independent","Top quark pole mass: 172.15 GeV at NNLO","Four PDF sets agree on top mass within 0.6 GeV","NNLO extraction rivals PDG top mass uncertainty"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"Everything rests on the assumption that the $q_T$-subtraction power corrections left out of the MATRIX predictions are small enough to be covered by the 1% uncorrelated theory uncertainty added per bin; if they are actually larger, or correlated with the top-pair invariant mass, the extracted pole mass would shift.","fun_headline_variants_meta":{"raw":{"variants":["NNLO fit pins top mass to 0.23 GeV precision","Top mass from NNLO QCD: PDF set independent","Top quark pole mass: 172.15 GeV at NNLO","Four PDF sets agree on top mass within 0.6 GeV","NNLO extraction rivals PDG top mass uncertainty"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000281,"raw_usage":{"total_tokens":1601,"prompt_tokens":819,"completion_tokens":782,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":435,"completion_tokens_details":{"reasoning_tokens":697}},"tokens_in":435,"tokens_out":782,"duration_ms":7806,"temperature":1.0,"reasoning_tokens":697,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-10T23:24:16.029623+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Take the exact set of normalized $M(t\\bar t)$ and double-differential bins used in the fit and compare MATRIX with an independent local-subtraction NNLO code such as STRIPPER at the same scales and mass values; if the bin-by-bin differences, when propagated through the $\\chi^2$ parabola fit, shift $m_t^{\\rm pole}$ by more than the quoted data uncertainty of roughly 0.2-0.3 GeV, then the power-correction assumption is falsified.","supporting_citations":[{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"The PDG 2024 review providing the preferred top-quark mass value that the fit results are compared with."}],"review_version":1}