{"id":"b8cad076-03e2-47a0-8c1e-e83897051e1c","arxiv_id":"2412.00350","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":4.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":3,"one_line_summary":"New LHC inclusive jet data push the CT18 gluon harder at high x and shrink its uncertainty, while dijet data show stronger scale dependence and are set aside.","lead":"This paper updates the CTEQ-TEA (CT18) picture of the proton by testing new LHC inclusive jet and dijet measurements at NNLO. It finds that inclusive jet data, unlike dijets, constrain the high-momentum gluon with little sensitivity to scale choices, and it releases the resulting CT18+nIncJet PDFs for future use.","discovery_kind":"extension","skeptic_critique":{"model":"deepseek-v4-flash","headline":"CT18+nIncJet rests on ePump profiling rather than a full CT18 refit; without a direct validation, the reported gluon shift and uncertainty reduction are provisional.","rationale":"Reader's weakest assumption correctly identifies the reliance on ePump, and I agree this is the load-bearing point. The central claim is not merely qualitative (new jets favor a harder gluon) but quantitative (PDF uncertainty reduced, cross-section shifts). These numbers are taken from CT18+nIncJet, which is generated by ePump profiling. ePump is a standard method for exploratory impact studies and is well suited to the dataset-selection parts of this paper; earlier CTEQ-TEA papers use it for similar updates. However, when the output is presented as a new PDF set with final phenomenological predictions, the equivalence to a full refit must be demonstrated. The manuscript contains an internal statement to this effect: Sec. V says ePump was employed \"before conducting time-intensive global fittings\", which is a limitation flag. The paper does compare chi2 values with MSHT and NNPDF, but those groups performed full fits; agreement in chi2 does not by itself prove agreement in PDFs. The proposed test is feasible because the CTEQ-TEA collaboration has the fitting machinery; a single full fit with three added datasets, or at least a release of the ePump-generated grids/PDF files for independent validation, would settle the question. Other concerns (data-driven decorrelation selection, scale choice) are secondary, because the authors show the PDFs are stable across options; the missing full-fit validation is the one that directly tests the main quantitative output.","tokens_in":33143,"tokens_out":4227,"duration_ms":40741,"concrete_test":"Run a full CT18 global fit with the three inclusive jet datasets (ATL8IncJet, ATL13IncJet, CMS13IncJet) added, using Method 5 theory treatment and the chosen decorrelations (SuggestedR6 for ATL8, Option18 for ATL13), and compare the resulting g(x, Q=100 GeV) central values and 90% C.L. bands at x = 0.05, 0.1, 0.2, 0.3, 0.5 with CT18+nIncJet. If the full-fit gluon central shift relative to CT18 differs from the ePump result by more than one-third of the ePump band width (or 1% in g), the CT18+nIncJet set should be labeled provisional until a full refit is released.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central claim that CT18+nIncJet is a reliable updated PDF set rests on ePump Hessian profiling rather than a full CT18 global refit. In Sec. III A the authors state \"we performed PDF fit using ePump\", and all CT18+nIncJet results in Sec. IV A (Fig. 10, Table VI) are produced by that method. ePump (Refs. [48,49]) reuses the baseline Hessian error sets and adds new data through a profiling chi-square; it is not equivalent in general to a new global minimization in which PDF parameters, tolerance, and correlations with all other datasets are refit. The paper gives no direct comparison of CT18+nIncJet with a full CT18 fit containing the same three inclusive jet datasets, so the approximation error is unknown. The conclusion even describes ePump as a step taken \"before conducting time-intensive global fittings\", indicating that the final set is not a full refit. Since the headline result is a shift and narrowing of the gluon PDF, a profiling bias of order 1% in g(x, Q=100 GeV) at x~0.1-0.4 would change the phenomenological conclusions in Sec. IV B.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper assesses the impact of recent LHC inclusive jet and dijet measurements on the CTEQ-TEA (CT18) global PDF analysis. The authors introduce an intermediate baseline, CT18mLHCJet, apply the ePump Hessian-profiling package to add new ATLAS and CMS jet datasets, and systematically examine decorrelation options for ATLAS systematic uncertainties, five treatments of NNLO theory uncertainties and K-factor smoothing, and two central scale choices with factor-of-two variations. They conclude that inclusive jet data have weaker scale dependence than dijet data, impose stronger constraints on the gluon PDF, and prefer a harder large-x gluon; this leads to the proposed CT18+nIncJet PDF set with reduced gluon uncertainties. The phenomenological section computes gluon-gluon luminosities and 14 TeV cross-sections for Higgs, top-quark pair, and ttH production using the updated PDFs.","tokens_in":33361,"tokens_out":5606,"duration_ms":56558,"significance":"If the ePump-based updating is a faithful proxy for a full global refit, the paper is a useful and timely impact study: it is transparent about the many methodological choices, compares with MSHT and NNPDF where data overlap, and provides detailed robustness checks across theory treatments, scales, and decorrelation options. The main findings—that new inclusive jet data pull the large-x gluon harder and reduce its uncertainty—are qualitatively consistent with other groups and are likely to inform the upcoming CTEQ-TEA global fit. However, since the central CT18+nIncJet set is produced by approximate Hessian profiling rather than a full CT18 refit, the quantitative claims of a gluon shift and uncertainty reduction remain provisional until validated against a complete refit.","major_comments":[{"comment":"The central result, CT18+nIncJet, is obtained with the ePump Hessian-profiling package and not with a full CT18 global refit. In Sec. III A the authors write 'we performed PDF fit using ePump', and in Sec. V they describe ePump as a step taken 'before conducting time-intensive global fittings'. ePump reuses the baseline Hessian error sets and applies a profiling chi-square; it is an approximation that can differ from a new global minimization in which PDF parameters, tolerance, and correlations with all other datasets are refit. Because the headline claims are a shift of roughly 1-3% in the large-x gluon and a reduction in its error band, an unvalidated profiling bias of comparable size would change the conclusions. The authors should either provide a direct comparison of CT18+nIncJet with a full CT18 fit containing the same three inclusive jet datasets, or explicitly reframe the paper as an ePump-based impact study whose quantitative PDF set is provisional.","section":"Sec. III, first paragraph"},{"comment":"The construction of the baseline CT18mLHCJet set is not fully specified. The text says it 'uses the same framework as the CT18 analysis' but excludes the inclusive jet datasets already in CT18, without stating whether this baseline was produced by a full global refit or by an ePump-based removal of datasets. This distinction matters because every subsequent ePump update is performed relative to CT18mLHCJet; if the baseline is itself an ePump approximation, then the final PDFs inherit that approximation in a way that is not assessed. Please state explicitly how CT18mLHCJet was created and, if it was a full fit, give the relevant fit quality and parameter shifts.","section":"Sec. III B, Figs. 6-8"},{"comment":"The preference for inclusive jet over dijet datasets is based on a qualitative comparison of PDF ratio plots and chi-square tables. For example, the text notes that for CMS8DiJet the central gluon PDF varies by roughly 4% at x ~ 0.3 under scale variations, while inclusive jet datasets vary by about 1-1.5%, but no quantitative criterion is given for declaring one set 'weakly scale dependent'. Since this preference is a central conclusion and also determines which datasets enter the final fit, the authors should define a quantitative measure of scale dependence (for instance, the maximum shift in g(x, Q = 100 GeV) over the fitted x-range under factor-of-two scale variations) and apply it uniformly to inclusive jet and dijet datasets.","section":"Sec. II and Sec. III A"},{"comment":"Several methodological choices are made after inspecting the data and are selected on the basis of the resulting chi-square: the ATLAS decorrelation options (SuggestedR6 and Option18), the theory treatment (Method 5), and the central scale (pT^jet for the final fit). This multiple-comparison selection, performed on the very data used to define CT18+nIncJet, is not reflected in the reported PDF uncertainties. The authors show that the PDF central values and error bands are largely insensitive to these choices, which mitigates the concern, but the conclusion that the new data reduce the gluon uncertainty would be strengthened by a discussion of how much of the improvement is a selection effect rather than a stable constraint from the data.","section":"Sec. IV B"}],"minor_comments":[{"comment":"The integrated luminosity for the CMS 13 TeV inclusive jet dataset is listed as 33.5 fb^-1 in Table I but 36.5 fb^-1 in Table VII; these should be reconciled.","section":"Table I vs Table VII"},{"comment":"The caption of Fig. 8 reads 'Same as Fig. 8, but for CMS8DiJet...'; it should refer to Fig. 7 (or to Fig. 6, depending on intent).","section":"Fig. 8 caption"},{"comment":"The experiment labels such as 'ATLAS8ZpT' and 'CMS8ttb' are introduced without definition; a short expansion or a pointer to the table of dataset IDs would improve readability.","section":"Fig. 12 and Sec. IV A"},{"comment":"The scale choice for ttH production is given as 'MT/2' without defining MT; please define it as the scalar sum of transverse masses of the final-state particles.","section":"Sec. IV B"},{"comment":"In the sentence 'In contrast, the ATLAS data at 13 TeV and 8 TeV show less significant deviations compared to the baseline', it would be clearer to list all datasets in the same order as the preceding ranking and to specify the x value at which the comparison is made.","section":"Appendix A 1"},{"comment":"The cross-sections in Fig. 15 and Fig. 16 are applications of the fitted PDFs rather than independent predictions; the text should state this explicitly so that readers do not interpret them as a posteriori validation of CT18+nIncJet.","section":"Sec. IV B"}],"recommendation":"major_revision","confidential_remarks":"The manuscript is honest about its use of ePump, and the robustness checks are a strength. The main issue is that the central PDF set is not validated against a full CT18 refit, which is essential before the quantitative gluon shift and uncertainty reduction can be taken at face value. I do not see a circularity problem beyond the standard fact that phenomenological applications of a fitted PDF are not independent tests."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Two things to know. First, this is a solid, transparent impact study: the authors systematically test decorrelation options, five theory treatments for NNLO K-factor MC errors, and scale choices, and they compare with MSHT and NNPDF. The central claim—that the newer ATLAS 8/13 TeV inclusive jet data pull the large-x gluon harder and shrink its uncertainty—matches what MSHT and NNPDF find, which gives me confidence the direction is right. Second, the paper is not a full CT18 refit. The CT18+nIncJet set is produced with ePump, a Hessian profiling shortcut, and the paper itself says the full fitting is still to come. That makes the set provisional; the size of the gluon shift and the error reduction could change once the parameters are truly refit.\n\nWhat's new: the CT18+nIncJet set itself, the scale-dependence comparison showing inclusive jets are more robust than dijets (with CMS 8 TeV dijets the worst offender), and the L2 sensitivity maps. The comparison table against MSHT and NNPDF chi-square values is useful. The phenomenological section is competently done—n3loxs, Top++, MadGraph—but it uses the same PDFs just fitted, so it's an application, not an independent prediction.\n\nSoft spots, in proportion. The ePump caveat is real but not disqualifying: ePump is established and prior work shows it tracks full fits reasonably, but this paper gives no direct validation, so the reader can't tell whether the ~1-2% gluon shift at x~0.1-0.4 is an artifact of profiling. The choice of decorrelation option and theory method is driven by best chi-square, which risks overfitting the new data; the authors do check stability across options and methods, which mitigates the concern, but not fully. No PDF files or code are released, which limits reproducibility. None of these are fatal; they just mean the set should be labeled as an update study, not a final PDF release.\n\nWho is this for? PDF phenomenologists and anyone who uses CT18 for LHC predictions. It deserves a serious referee. I'd send it to review and ask the authors to either validate ePump against a full refit or clearly frame CT18+nIncJet as an intermediate update.","headline":"A careful, useful impact study of new LHC jet data on CT18, with the right caveat: the headline gluon shift rests on ePump profiling rather than a full refit, so treat CT18+nIncJet as provisional even though the direction is consistent with MSHT and NNPDF.","tokens_in":33951,"tokens_out":2265,"would_cite":true,"duration_ms":21512,"reading_group":"yes","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":[],"pacs":[],"model":"deepseek-v4-flash","headline":"The inclusion of the latest LHC inclusive jet measurements in the CT18 global PDF analysis reduces gluon uncertainties and pushes the large-x gluon harder, and the paper argues inclusive jet data should be preferred over dijet data…","keywords":["parton distribution functions","gluon PDF","inclusive jet production","dijet production","NNLO QCD","scale dependence","Hessian profiling","LHC phenomenology"],"falsifier":"Run a full CT18 global fit that includes the same three inclusive jet datasets, using the same smoothing and Monte Carlo integration-error treatment, and compare the large-$x$ gluon PDF with CT18+nIncJet. If the full fit's central gluon at $x\\approx 0.3$ falls outside the reduced uncertainty band claimed here, the ePump approximation is not reproducing the global fit and the paper's quantitative claims would not survive.","tokens_in":32919,"feed_emoji":"⚛️","tokens_out":8765,"duration_ms":76647,"temperature":0.7,"pith_summary":"New LHC measurements of inclusive jet and dijet production, analyzed with next-to-next-to-leading-order QCD predictions, can sharpen what we know about the gluon inside the proton. This paper argues that the inclusive jet datasets are the better tool for that job: their predicted cross sections barely change when the renormalization and factorization scales are varied, while dijet predictions, especially those from the 8 TeV dijet data, shift strongly. Adding three inclusive jet datasets to the baseline CT18 global fit yields a new proton PDF set in which the gluon at large momentum fraction $x$ is harder and carries a smaller uncertainty. If the claim holds, the updated set is a more reliable input for gluon-dominated LHC processes such as Higgs, top-pair, and $t\\bar{t}H$ production.","feed_headline":"New jet data harden the proton's high-x gluon","feed_subtitle":"Inclusive jet measurements shrink gluon error bands and sharpen Higgs, top-pair, and ttH predictions at the LHC.","key_machinery":"The study is carried by a three-part apparatus. Precomputed NNLO interpolation grids supply the jet cross-section predictions, with renormalization/factorization scales $p_T^{\\mathrm{jet}}$ or $\\hat H_T$ for inclusive jets and $m_{12}$ or $p_{T,1}e^{0.3y^*}$ for dijets; the ePump Hessian profiling procedure then approximates how each new dataset shifts the best-fit PDFs and their Hessian error bands without a full refit; and the L2 sensitivity quantifies, at each $x$, how strongly each experimental dataset pulls the gluon PDF. The scale-dependence comparison is the argument's decisive test: it is what separates 'safe' inclusive jet data from 'risky' dijet data, particularly the 8 TeV dijet measurement.","core_discovery":"The paper's central claim is that the latest inclusive jet data from the LHC, once included in the CT18 analysis through a fast Hessian profiling update, reduce the gluon PDF uncertainty and move the central gluon distribution upward for $x\\gtrsim 0.1$, producing a new set called CT18+nIncJet. A second, distinct claim is that inclusive jet data should be preferred over dijet data for constraining the gluon at this stage: the inclusive jet predictions are insensitive to the choice and variation of the central renormalization and factorization scales, whereas dijet predictions, most notably the 8 TeV dijet data, show a strong scale dependence that can change the central gluon by several percent. The paper also shows that inclusive jet datasets pull the gluon PDF more strongly than the dijet datasets from the same collisions, and that the new set raises the gluon-gluon luminosity at large invariant masses, increasing high-mass Higgs-like scalar, top-pair, and $t\\bar{t}H$ cross sections at 14 TeV while slightly lowering the inclusive 125 GeV Higgs cross section.","pith_inferences":["If a future full refit confirms the harder high-$x$ gluon, the same shift should appear in gluon-dominated observables not refit here, such as high-$p_T$ photon-plus-jet production; a consistent shift there would be an independent check.","The paper's preference for inclusive jets is based on scale dependence, not fit quality, since several dijet datasets fit better. A reader should therefore expect future PDF releases to omit dijets until their scale ambiguity is resolved; a triple-differential dijet measurement with scale behavior similar to inclusive jets would be the natural test.","Because the numerical size of the uncertainty reduction rests on the ePump approximation, the quoted error bands of CT18+nIncJet should be treated as provisional until a full refit validates them."],"forward_implications":["The gluon PDF uncertainty shrinks, with the largest reduction at large $x$, and the central gluon becomes harder.","The gluon-gluon luminosity $L_{gg}$ is enhanced at large invariant masses with narrower error bands.","Higgs-like scalar production at masses above roughly 1 TeV increases, while the inclusive 125 GeV Higgs cross section decreases slightly.","Top-pair and $t\\bar tH$ cross sections at 14 TeV increase mildly, reflecting their positive correlation with the large-$x$ gluon.","Inclusive jet data are preferred over dijet data for future PDF fits, and dijet data, particularly the 8 TeV dijet set, require further theoretical work on scale dependence before inclusion."],"supporting_citations":[{"why":"Defines the CT18 NNLO baseline PDF set and tolerance criterion that the new datasets update.","marker":"[1]"},{"why":"Supplies the NNLO interpolation grids for inclusive jet and dijet cross sections used in all predictions.","marker":"[41]"},{"why":"Introduces the Hessian profiling method used to update PDFs with new data in this analysis.","marker":"[48]"},{"why":"Extends and benchmarks the Hessian updating method used throughout the paper.","marker":"[49]"},{"why":"Provides the 8 TeV inclusive jet dataset that is a major driver of the harder high-$x$ gluon.","marker":"[23]"},{"why":"Provides the 13 TeV inclusive jet and dijet dataset, another main constraint on the large-$x$ gluon.","marker":"[25]"},{"why":"Provides the 13 TeV inclusive jet dataset whose pull is somewhat weaker but still included in the combined set.","marker":"[27]"},{"why":"Supplies the comparison of jet data impact and fit quality from an independent global analysis.","marker":"[70]"},{"why":"Supplies the comparison of inclusive jet and dijet constraints and scale behavior from an independent global analysis.","marker":"[72]"}],"fun_headline_variants":["Inclusive jets sharpen gluon, reframe LHC cross sections","LHC inclusive jets beat dijets for gluon precision","Tighter gluon from inclusive jets shifts Higgs and top rates","CT18 update: inclusive jets dominate gluon constraints"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"Everything rests on the fast Hessian-profiling update reproducing what a full CT18 global refit with the same datasets would produce, since all of the paper's new PDFs, uncertainties, and sensitivity plots come from that approximate update and no direct full-refit comparison is shown.","fun_headline_variants_meta":{"raw":{"variants":["Inclusive jets sharpen gluon, reframe LHC cross sections","LHC inclusive jets beat dijets for gluon precision","Tighter gluon from inclusive jets shifts Higgs and top rates","CT18 update: inclusive jets dominate gluon constraints"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000366,"raw_usage":{"total_tokens":1962,"prompt_tokens":931,"completion_tokens":1031,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":547,"completion_tokens_details":{"reasoning_tokens":962}},"tokens_in":547,"tokens_out":1031,"duration_ms":10535,"temperature":1.0,"reasoning_tokens":962,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-12T05:28:00.775152+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Run a full CT18 global fit that includes the same three inclusive jet datasets, using the same smoothing and Monte Carlo integration-error treatment, and compare the large-$x$ gluon PDF with CT18+nIncJet. If the full fit's central gluon at $x\\approx 0.3$ falls outside the reduced uncertainty band claimed here, the ePump approximation is not reproducing the global fit and the paper's quantitative claims would not survive.","supporting_citations":[{"cited_title":"Impact of the LHCb 8 W, Z, and asymmetry data on the CT14HERA2 PDFs,","cited_arxiv_id":null,"evidence_quote":"Supplies the comparison of jet data impact and fit quality from an independent global analysis."}],"review_version":1}