{"id":"91c923dd-c7f4-4abf-8499-51232556030a","arxiv_id":"2412.15742","paper_version":2,"verdict":"ACCEPT","confidence":"HIGH","novelty_score":6.0,"correctness_risk":"low","formal_verification":"none","parameter_count":1,"one_line_summary":"ATLAS reports the first differential D_s± meson cross-section measurement and extends D±/D_s± data to pT = 100 GeV at 13 TeV, finding consistency with NLO QCD within uncertainties.","lead":"The ATLAS experiment measured how often D± and D_s± charmed mesons are produced in high-energy proton collisions, over a wider momentum range than before. The new data, from 137 inverse femtobarns of LHC collisions, stress-test quantum chromodynamics and can calibrate future particle searches.","discovery_kind":"extension","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The acceptance-efficiency corrections C_i/C_j are derived from Pythia8 LO MC reweighted to the reconstructed pT spectrum; without a reported closure test, a residual MC-shape bias in the bin-by-bin corrections cannot be excluded, which directly affects the central cross-section values.","rationale":"The reader's weakest assumption identified the MC-based acceptance and efficiency corrections as the main risk to the central claim; I agree and sharpen it to a specific, testable gap: the pT reweighting is performed on the reconstructed spectrum, and no closure test is reported to verify that the bin-by-bin correction procedure is unbiased for true spectra differing from Pythia8. The high-pT discrepancy for the D_s meson (80-100 GeV bin, about 4 sigma from GM-VFNS) is noteworthy but concerns the consistency statement in the abstract rather than the measurement itself; the paper's own conclusion already acknowledges a high-pT deviation. The measurement is otherwise well-documented, with comprehensive systematics and internal consistency checks (e.g., the pT and |eta| sums agree). Because the acceptance-correction concern is not confirmed by concrete evidence, the reader's ACCEPT verdict stands, but a closure test would settle whether the concern actually lands.","tokens_in":54395,"tokens_out":15974,"duration_ms":141346,"concrete_test":"Run a full-simulation closure test: generate pseudo-data from a true D-meson pT spectrum that deliberately differs from the Pythia8 LO prediction (for example, a spectrum hardened or softened by ±20% in slope, or taken from the GM-VFNS shape), pass it through the same reconstruction and selection, apply the same pT and f_NP reweighting to the MC, compute the C_i corrections, and compare the extracted cross-sections to the known input. If the residual bias exceeds the quoted 'Kinematics' systematic uncertainty in any pT bin, the MC-dependence of the acceptance corrections is underestimated and the central values need revision.","verdict_should_be":"UNCHANGED","load_bearing_attack":"Equation 4 divides the extracted yields by bin-by-bin acceptance-efficiency factors C_i and C_j computed from Pythia8 LO MC events. The MC pT spectrum is reweighted to match the distribution extracted from data (Section 6, 'Kinematics'), and the non-prompt fraction is reweighted to the f_NP from the lifetime fit. This reweighting targets the reconstructed spectrum, which is already shaped by the same efficiency; within a bin, particularly the 12-15 GeV bin where the trigger efficiency turns on steeply, the average efficiency is sensitive to the assumed true pT shape. The 'Kinematics' systematic propagates only the statistical uncertainty of the data spectrum, not a systematic difference between the reweighted MC shape and the true underlying shape. If the MC within-bin shape is wrong, C_i and C_j are biased and all differential cross-sections shift coherently. No closure test on simulation is reported that would demonstrate the correction procedure recovers a known input spectrum. This is the most load-bearing concern because it bears directly on the central values of the measurement, not just on the theory comparison.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"This paper reports measurements of the inclusive and differential production cross-sections of D± and D_s± mesons at sqrt(s)=13 TeV using 137 fb^-1 of ATLAS data, reconstructed in the decay channel D±/D_s± -> phi(mu mu) pi± in the range 12 < pT < 100 GeV and |eta| < 2.5. The yields are extracted from unbinned fits to the mu-mu-pi invariant mass, and the non-prompt fraction is constrained by a pseudo-proper-lifetime template fit. The differential cross-sections as functions of pT and |eta| are compared with GM-VFNS and FONLL predictions, and fiducial inclusive cross-sections are given for several pT thresholds. The central claim is that the NLO predictions are consistent with the data within the large theoretical uncertainties, with the caveat of a trend toward a high-pT deviation.","tokens_in":54555,"tokens_out":11221,"duration_ms":101037,"significance":"If the measurement is correct, this is the first ATLAS differential D_s± cross-section measurement and the first LHC D_s± measurement up to pT=100 GeV, extending the kinematic reach of charm-production data in a region where GM-VFNS and FONLL predictions differ non-negligibly. The paper is careful to quote statistical, systematic, and branching-ratio uncertainties separately and to document the fit models and the propagation of the non-prompt fraction uncertainty. These features make the result useful for testing heavy-quark production calculations and fragmentation-function inputs, even though no public code or closure-test suite is provided.","major_comments":[{"comment":"The correction factors C_i and C_j in Eq. (4) are computed bin-by-bin from Pythia8 LO MC events whose pT spectrum is reweighted to the distribution extracted from data (Section 6, 'Kinematics'), and the quoted systematic propagates only the statistical uncertainty of the data spectrum. Because the average efficiency in a bin is sensitive to the assumed true pT shape, and the 12–15 GeV bin lies in the steep trigger-efficiency turn-on, a residual difference between the reweighted MC shape and the true shape would bias all C_i and hence all central cross-section values. No closure test is reported that demonstrates the correction procedure recovers a known input spectrum. Please add such a closure test (or an alternative efficiency estimate from a second generator) and/or a shape systematic that covers the within-bin model dependence.","section":"§5.3 Eq. (4), §6 Kinematics"},{"comment":"The abstract states that the predictions 'are found to be consistent with the measurements in the visible kinematic region within the large theoretical uncertainties,' but Table 5 shows the D_s± 80<pT<100 GeV bin at 8.1±0.9(stat)±0.9(syst)±0.6(BR) compared with GM-VFNS 13.8(+0.8,−1.1), a discrepancy of roughly 3 standard deviations. The conclusion's 'slightly deviation towards high-pT regions' is closer to the data. Please qualify the consistency claim to reflect the high-pT trend, or discuss why this bin should not be interpreted as a tension.","section":"Abstract and §7, Table 5"},{"comment":"The non-prompt fraction f_NP is obtained from a template fit in which all shape parameters of the prompt and non-prompt lifetime templates are fixed to MC values (Eq. (3)), and Section 6 propagates only the statistical uncertainty of f_NP to the acceptance corrections. No systematic is assigned for the template shapes themselves (e.g., the B-meson lifetime component, the resolution smearing, or the turn-on parameters), although a bias in f_NP would alter the prompt/non-prompt mixture used to reweight the MC and thus C_i/C_j. Please include a template-shape systematic or demonstrate quantitatively that the resulting change in the correction factors is negligible.","section":"§5.2 Eq. (3), §6 Non-prompt"}],"minor_comments":[{"comment":"The sentence 'The statistical uncertainties for D± mesons are larger than those for D_s± mesons. This is due to the lower yield ...' is inconsistent with the yields in Fig. 1 (D±: 68350, D_s±: 21780); the lower-yield explanation should be corrected or removed.","section":"§5.2 (text after Fig. 3)"},{"comment":"The description of the reweighting is ambiguous: it is not stated whether the data spectrum used for reweighting is the reconstructed spectrum or an efficiency-corrected and unfolded spectrum. Please state this explicitly in the text.","section":"§6 Kinematics"},{"comment":"The phrasing 'as the uncertainties of the world average branching ratios ... combined are better than the branching ratio of D_s±→φπ±' is awkward and should be reworded for clarity.","section":"Eq. (5)"},{"comment":"In the ratio panels, the theory uncertainties are shown as hatched bands but it is not stated whether the data points include statistical only or total uncertainties; please clarify in the caption.","section":"Figure 8 caption"}],"recommendation":"major_revision","confidential_remarks":"The measurement appears carefully executed and the concerns above are documentation- and presentation-level in nature: adding a closure test, clarifying the reweighting, and softening the abstract claim would address the load-bearing issues. I do not see grounds for rejection."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"This is a genuine new measurement: first ATLAS D_s± differential cross-section, and first at any LHC experiment up to pT = 100 GeV. The decay channel D→φ(μμ)π with a di-muon trigger is an original way to access charm at high pT, and the analysis is thorough: separate statistical, systematic, and branching-ratio uncertainties, detailed fit models, and comparison with GM-VFNS and FONLL.\n\nThe central result holds up. Data and predictions agree within the large theory uncertainties over most of the range; the GM-VFNS sits high at large pT, especially for D_s±, but the uncertainties are big enough that this is a mild trend, not a discrepancy. The tables are internally consistent, and the fiducial cross-section sums over pT and |η| agree.\n\nThe weakest point is exactly what the stress-test flags: the C_i and C_j corrections come from Pythia8 LO MC, reweighted to the reconstructed data pT and f_NP. The 'Kinematics' systematic propagates only the statistical uncertainty of that data spectrum, not a systematic uncertainty on the within-bin shape. For the 12–15 GeV bin, where the trigger efficiency is steep, the average efficiency could be sensitive to the assumed true shape. The paper does not report a closure test on simulation that would show the procedure recovers a known input spectrum. That does not invalidate the measurement — this is standard practice in the field — but it means the central values carry an unquantified modeling component that is not in the quoted systematics.\n\nAlso, for D± the background-model uncertainty dominates and drives total systematics of 15–20%, which is large but honestly reported. The BR uncertainties are ~7% and are what they are.\n\nThis paper is for heavy-flavor phenomenologists and for anyone needing a high-pT charm cross-section normalization. It deserves a serious referee; the analysis is documented well enough to check, and the result fills a real gap.\n\nI would send it to peer review with a request that the authors either add a closure test for the efficiency corrections or explicitly discuss the sensitivity of the first pT bin to the MC shape. That is a minor addition; the measurement itself is sound.","headline":"A solid new ATLAS charm-production measurement that deserves refereeing; the main caveat is the MC-derived acceptance corrections, which would benefit from a closure test.","tokens_in":55190,"tokens_out":2382,"would_cite":true,"duration_ms":25064,"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 measures inclusive $D^\\pm$ and $D_s^\\pm$ meson production cross-sections in 13 TeV proton–proton collisions for $12 < p_{\\mathrm{T}} < 100$ GeV and $|\\eta| < 2.5$, and finds them consistent with next-to-leading-order QCD…","keywords":["D± meson","D_s± meson","charm production cross-section","13 TeV proton–proton collisions","general-mass variable-flavour-number scheme","FONLL","non-prompt fraction","fiducial cross-section"],"falsifier":"Measure the same fiducial cross-sections in the same kinematic region using independent hadronic decay channels, such as $D^\\pm\\to K^\\mp\\pi^\\pm\\pi^\\pm$ and $D_s^\\pm\\to\\phi(K^+K^-)\\pi^\\pm$, or with separated prompt and non-prompt samples; a disagreement larger than the quoted systematic uncertainties would show that the semileptonic acceptance corrections or the lifetime-template reweighting are biased.","tokens_in":54126,"feed_emoji":"⚛️","tokens_out":11010,"duration_ms":84265,"temperature":0.7,"pith_summary":"The paper measures how often $D^\\pm$ and $D_s^\\pm$ charmed mesons are produced in 13 TeV proton–proton collisions, using the decay chains $D^\\pm/D_s^\\pm \\to \\phi(\\mu\\mu)\\pi^\\pm$ and reconstructing a two-muon-plus-pion vertex. It reports inclusive (prompt plus non-prompt) differential cross-sections in nine bins of transverse momentum from 12 to 100 GeV and five bins of absolute pseudorapidity below 2.5, based on 137 fb$^{-1}$ of data. The central claim is that the measurements are consistent with next-to-leading-order QCD predictions within the large theoretical uncertainties, with the GM-VFNS prediction tending to lie above the data at high $p_{\\mathrm{T}}$. If this is right, the results provide the first differential $D_s^\\pm$ measurement from this experiment and the first $D_s^\\pm$ data reaching 100 GeV at the LHC, a benchmark for charm-production calculations in an unexplored kinematic range.","feed_headline":"D± and D_s± meson rates reach 100 GeV, matching QCD","feed_subtitle":"First differential D_s± measurement at 13 TeV gives a benchmark for QCD predictions up to 100 GeV.","key_machinery":"The load-bearing object is the semileptonic decay chain $D^\\pm/D_s^\\pm \\to \\phi(\\mu\\mu)\\pi^\\pm$, whose clean two-muon signature suppresses the large combinatorial background from hadronic activity. Signal yields come from a simultaneous unbinned fit to the $m_{\\mu\\mu\\pi}$ invariant-mass distribution, with non-relativistic Voigtian peaks for the two mesons and a normalised quadratic-exponential background; the mass difference $m_{D_s^\\pm} - m_{D^\\pm}$ is constrained to its world-average value. The prompt/non-prompt composition is fixed by a template fit to the pseudo-proper lifetime $\\tau = m_{\\mu\\mu\\pi}\\,L_{xy}/p_{\\mathrm{T}}$, using convolution templates built from exponential, Gaussian, and error-function terms for charm and bottom decays. The cross-section in each bin is $d\\sigma/dp_{\\mathrm{T}} = S_i / (\\int\\!\\mathcal{L}\\,dt\\, C_i\\, \\mathcal{B}\\, \\Delta_i p_{\\mathrm{T}})$, where $S_i$ is the fitted yield, $C_i$ is the MC-derived acceptance-times-efficiency correction reweighted to data kinematics, $\\mathcal{B}$ is the decay branching ratio, and $\\Delta_i p_{\\mathrm{T}}$ is the bin width.","core_discovery":"The paper establishes that charmed-meson production in the visible region $12 < p_{\\mathrm{T}} < 100$ GeV, $|\\eta| < 2.5$ is described by next-to-leading-order QCD within uncertainties. Signal yields for the two mesons are extracted simultaneously by an unbinned maximum-likelihood fit to the $m_{\\mu\\mu\\pi}$ spectrum, with Voigtian signal shapes, a quadratic-exponential background, and the $D_s^\\pm - D^\\pm$ mass difference constrained by a Gaussian penalty. The conversion from yields to cross-sections applies bin-by-bin acceptance and efficiency corrections computed from simulated events, with the simulation reweighted to match the measured non-prompt fraction and the data $p_{\\mathrm{T}}$ spectrum. The fiducial cross-section for $D^\\pm$ in $12 < p_{\\mathrm{T}} < 100$ GeV, $|\\eta|<2.5$ is measured at about 10.8 $\\mu$b with total uncertainties near 14–20 percent, and for $D_s^\\pm$ at about 5.0 $\\mu$b; GM-VFNS agrees for both, FONLL agrees for $D^\\pm$, and no FONLL prediction exists for $D_s^\\pm$.","pith_inferences":["Because many scale and PDF uncertainties cancel, the $D_s^\\pm/D^\\pm$ cross-section ratio as a function of $p_{\\mathrm{T}}$ would test fragmentation and strangeness-suppression effects more sharply than the absolute rates, and the paper's tables make that ratio directly computable.","If the GM-VFNS high-$p_{\\mathrm{T}}$ tendency is real rather than a scale-uncertainty artefact, it would point to missing physics in the massive- versus massless-scheme matching; a future FONLL prediction for $D_s^\\pm$ would settle whether the same trend appears in both schemes.","The non-prompt fractions extracted here carry large uncertainties and are explicitly not compared with theory; a dedicated prompt/non-prompt separation in the same kinematic range would show whether the inclusive agreement hides compensating discrepancies between charm and bottom contributions."],"forward_implications":["The $D_s^\\pm$ differential cross-section becomes available for the first time from this experiment, and $D^\\pm$ and $D_s^\\pm$ data now extend to $p_{\\mathrm{T}} = 100$ GeV, beyond the reach of earlier $D_s^\\pm$ measurements.","The agreement with GM-VFNS and FONLL gives a quantitative validation of next-to-leading-order charm production in the visible region, while the high-$p_{\\mathrm{T}}$ tendency of GM-VFNS to sit above the data identifies where scale and fragmentation uncertainties should be revisited.","The measured 13-to-7 TeV ratio of fiducial cross-sections is consistent with both theoretical predictions, supporting the predicted energy scaling of charm production.","The published cross-sections provide a normalisation for physics analyses that count heavy-hadron decays as signal or background, including searches for lepton-flavour-violating $\\tau$ decays."],"supporting_citations":[{"why":"Earlier measurement of $D^{*\\pm}$, $D^\\pm$, and $D_s^\\pm$ production at 7 TeV; supplies the comparison baseline and the prior method setup.","marker":"[16]"},{"why":"Prior measurement of $D_s^\\pm$ production at 13 TeV up to $p_{\\mathrm{T}}=36$ GeV; defines the previous kinematic reach that this result extends.","marker":"[6]"},{"why":"Prior measurement of $D^\\pm$ production at 13 TeV up to $p_{\\mathrm{T}}=100$ GeV; provides the previous reach for the charged meson.","marker":"[12]"},{"why":"Establishes the GM-VFNS calculation used for both $D^\\pm$ and $D_s^\\pm$ comparisons.","marker":"[1, 32–35]"},{"why":"Establishes the FONLL calculation used for the $D^\\pm$ comparison; the prediction is not available for $D_s^\\pm$.","marker":"[36, 37]"},{"why":"Pythia8 event generator supplies the leading-order simulation from which acceptance and efficiency corrections are derived.","marker":"[25]"},{"why":"Luminosity determination that fixes the overall normalisation of the cross-sections at 137 fb$^{-1}$.","marker":"[24]"},{"why":"World-average branching ratios and mass difference used in the cross-section formula and the signal fit constraint.","marker":"[3]"}],"fun_headline_variants":["Charmed meson production matches QCD predictions at 13 TeV","First differential D_s± measurement at 13 TeV from ATLAS","D± and D_s± rates consistent with QCD from 12 to 100 GeV","ATLAS charmed meson rates agree with QCD predictions","Benchmark D_s± production in pp collisions at 13 TeV"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The whole correction chain stands on the simulated events used to compute acceptance and efficiency: if, after reweighting, they do not reproduce the real detector response and the true mix of promptly produced and bottom-decay-produced $D$ mesons, the measured cross-sections are biased by more than the quoted uncertainties.","fun_headline_variants_meta":{"raw":{"variants":["Charmed meson production matches QCD predictions at 13 TeV","First differential D_s± measurement at 13 TeV from ATLAS","D± and D_s± rates consistent with QCD from 12 to 100 GeV","ATLAS charmed meson rates agree with QCD predictions","Benchmark D_s± production in pp collisions at 13 TeV"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000652,"raw_usage":{"total_tokens":3016,"prompt_tokens":1002,"completion_tokens":2014,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":618,"completion_tokens_details":{"reasoning_tokens":1916}},"tokens_in":618,"tokens_out":2014,"duration_ms":13974,"temperature":1.0,"reasoning_tokens":1916,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-11T11:08:11.632068+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Measure the same fiducial cross-sections in the same kinematic region using independent hadronic decay channels, such as $D^\\pm\\to K^\\mp\\pi^\\pm\\pi^\\pm$ and $D_s^\\pm\\to\\phi(K^+K^-)\\pi^\\pm$, or with separated prompt and non-prompt samples; a disagreement larger than the quoted systematic uncertainties would show that the semileptonic acceptance corrections or the lifetime-template reweighting are biased.","supporting_citations":[{"cited_title":"Navas et al.,Review of particle physics, Phys","cited_arxiv_id":null,"evidence_quote":"World-average branching ratios and mass difference used in the cross-section formula and the signal fit constraint."}],"review_version":1}