{"id":"f5a0df3f-2d24-4eb6-89ec-40413ba272c2","arxiv_id":"1909.00452","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":2,"one_line_summary":"The paper introduces nCTEQ++, a new C++ framework for nuclear PDF fits, and reports a preliminary fit in which LHC pPb W/Z data can be accommodated by raising the strange quark PDF, though the effect is sensitive to normalization shifts.","lead":"The nCTEQ collaboration presents nCTEQ++, a modular C++ code for fitting nuclear parton distributions, and uses it to include LHC proton-lead W/Z data in a preliminary fit. The new data improve the fit, but the shift in the strange quark PDF depends strongly on how much normalization freedom the data sets are allowed.","discovery_kind":"new_method","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Restricted 9-parameter basis for non-strange flavors may force the 40% strange shift; test with full basis.","rationale":"The reader correctly identified the fragility of the 40% strange shift, emphasizing normalization shifts as the weakest assumption. However, a more fundamental confound is the reduced parameterization of the non-strange distributions: the fit moves from 16 to 9 parameters for {g,uV,dV,ubar+dbar} while freeing strange. This means the fit cannot explore previously allowed non-strange directions, so the least constrained flavor (strange) may absorb tensions that are actually due to other flavors. This concern is internal to the fitting setup and does not rely on the legitimacy of normalization shifts. The paper's own caveat ('we free up only a limited set of parameters') is explicitly flagged in the manuscript, and the conclusion says a complete set of free parameters is needed. Therefore the central numerical claim, the 40% increase, lacks a demonstration that it is driven by genuine data sensitivity. The reader's CONDITIONAL verdict remains appropriate: the framework is plausible and useful, but the physical claim needs a full-parameter fit with uncertainty propagation. My proposed test directly addresses this by re-fitting with the full non-strange basis and by checking against the nCTEQ15 uncertainty band. I do not see grounds to reject or unverify outright, because the paper is honest about its preliminary nature and the new nCTEQ++ framework is a genuine development.","tokens_in":5569,"tokens_out":5486,"duration_ms":55653,"concrete_test":"Refit the same pPb W/Z + DIS + DY data with the same settings but with the full nCTEQ15 16-parameter non-strange basis plus 3 strange parameters, and no normalization shifts. If the strange PDF shift at x~10^-2 and Q=2 GeV drops below ~20% or changes sign, the 40% claim is an artifact of the restricted fit. As an independent check, compute the nCTEQ15 1-sigma uncertainty band for s+sbar at the same x and Q; if the 40% shift lies within that band, the increase is not statistically significant.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central claim is the ~40% increase of the lead strange PDF at x relevant for pPb W/Z production when no normalization shifts are allowed. The fit in Section 4 uses only 12 free parameters: 3 for s+sbar and 9 for {g,uV,dV,ubar+dbar}, whereas nCTEQ15 uses 16 parameters for the non-strange set with strange fixed. This reduction from 16 to 9 parameters for the non-strange distributions means the fit cannot adjust several previously free directions (e.g., in gluon or sea shapes). If the LHC W/Z data prefer changes in those directions, the fit may push the strange parameters instead, since strange is the least constrained flavor introduced here. The paper itself asks whether the increase is nature or the fit exploiting an unconstrained flavor (Section 4), but it does not test the contribution of the reduced parameterization to this effect. The normalization-shift sensitivity (the reader's weakest assumption) is real: the shift is roughly halved at 1 sigma. But even the no-shift result is confounded by the restricted basis. Thus the 40% figure is not demonstrated to be a property of the data; it is a property of the fitting setup.","agreement_with_reader":"partial"},"referee_report":{"model":"deepseek-v4-flash","summary":"This proceedings contribution from the nCTEQ collaboration describes a new C++ implementation of the nuclear PDF fitting framework (nCTEQ++) that integrates HOPPET, APPLgrid, and MCFM, and its first application: a preliminary NLO global fit that adds LHC pPb W±/Z vector-boson production data to the existing DIS and Drell-Yan data sets of nCTEQ15. The fit uses 12 parameters, three of which describe the strange-plus-antistrange distribution. The central phenomenological result is that, when no normalization shifts are allowed for the LHC data, the fit raises the lead strange PDF by about 40% at the x values probed by W/Z production; allowing normalization shifts of up to 1σ roughly halves this shift. The paper explicitly asks whether the strange increase is dictated by nature or is an artifact of fitting one of the least constrained flavors, and states that a complete analysis with additional parameters and data is in progress.","tokens_in":5786,"tokens_out":3507,"duration_ms":34044,"significance":"If the reported strange-PDF shift were robust, it would be an interesting hint that LHC pPb W/Z data, together with nuclear corrections, can discriminate strange flavor in a way complementary to neutrino-DIS dimuon production. The paper has real strengths: it introduces a modular, maintainable fitting code (nCTEQ++) with validated APPLgrid-based NLO calculations, and it is admirably candid in identifying the normalization-shift dependence and the question of whether s(x) is simply being exploited as a least-constrained flavor. However, the analysis is explicitly preliminary and the central 40% figure is not accompanied by an uncertainty estimate, is obtained from a deliberately restricted parameter basis, and depends strongly on the normalization treatment. The significance of the claim is therefore conditional: the framework and the questions posed are valuable, but the quantitative conclusion is not yet established.","major_comments":[{"comment":"The 40% increase of the lead strange PDF is quoted as a single number with no uncertainty. The paper advertises nCTEQ nPDFs with uncertainties, and the nCTEQ15 baseline has an uncertainty band; the fitted s(x) should be compared with that band at the relevant x values. Without such a comparison, the reader cannot tell whether a 40% shift is statistically significant or consistent with the prior uncertainty, and the abstract and conclusion currently present the number without this qualification.","section":"Sec. 4, Fig. 4"},{"comment":"The restricted parameterization is a load-bearing confound for the central claim. The fit uses 12 parameters (3 for s+sbar and 9 for the remaining flavors), whereas nCTEQ15 uses 16 parameters for the non-strange set. Because the non-strange directions that are frozen here could absorb some of the LHC W/Z data's pull, the 40% strange shift may be an artifact of the reduced basis rather than a property of the data. The paper itself voices this concern in Sec. 4, but it does not provide a test, such as a fit with the full 16-parameter non-strange basis or a fit that frees the gluon and strange parameters separately, to determine how much of the shift survives.","section":"Sec. 4, footnote 3"},{"comment":"The no-normalization fit, which produces the 40% strange shift, has an overall chi2/dof of 828/816, but the figure shows that several individual W/Z data sets have unacceptable chi2 values. The paper notes that allowing up to 1-sigma normalization shifts reduces the strange shift by roughly half. Since the headline number is taken from a fit that does not adequately describe the new data, the central claim is not robust to the normalization treatment; at minimum, the paper should give the shift with a systematic uncertainty reflecting the choice of whether and how normalization shifts are allowed.","section":"Sec. 4, Fig. 3"}],"minor_comments":[{"comment":"The abstract uses the word “impact” for the effect of the LHC W/Z data, but the reported chi2 numbers and PDF shifts are results of a fit that includes those data, not predictions made before the fit; a qualifier noting that the result is a preliminary fit would be appropriate in the abstract.","section":"Abstract"},{"comment":"There is a typo in the abstract, “strucure” for “structure,” and in Sec. 2 “Y AML” appears where “YAML” is intended; these should be corrected.","section":"Sec. 2"},{"comment":"The caption contains the typos “Prelimnary” and “unconstrainted,” and the phrase “w/o Norm Penalty” should be explained, since the body text describes the shifts as constrained by luminosity uncertainties rather than by a penalty term.","section":"Fig. 3 caption"},{"comment":"The status table lists CMS Z (6235) and CMS_II W± (6232, 6234) as “grids finished not validated,” while the text says several data sets are “fully included”; the paper should specify explicitly which of the listed LHC data sets enter the reported fit and which are only shown as comparisons.","section":"Sec. 4, status table"},{"comment":"The vertical magenta line is described only as “the central x value for pPb W±/Z production”; stating the x range or value, and the Q scale used for the PDFs, would make the plot more informative.","section":"Fig. 4"}],"recommendation":"major_revision","confidential_remarks":"This is a proceedings contribution and the analysis is explicitly preliminary, so the appropriate standard may be more lenient than for a full research article. However, the central claim (the 40% strange shift) is not yet demonstrated: it lacks an uncertainty, is confounded by the reduced parameter basis, and depends on the normalization treatment. The authors' own caveats are to their credit, but the abstract and conclusion overstate the result relative to what is shown. If the claims are softened to match the presented evidence, or the fit is extended to include the full parameter basis and uncertainty bands, the paper would be publishable as a proceedings contribution. No other concerns about novelty or citation practice."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"The one thing to know: this is an honest progress report, not a physics result. The nCTEQ collaboration has built a new C++ harness (nCTEQ++) that can include LHC W/Z observables in nuclear PDF fits, and they've run a first, deliberately restricted fit with pPb data. The eye-catching 40% rise in the lead strange PDF is fragile: it depends on both the fit's limited parameter count and on whether normalization shifts are allowed. The authors say this themselves, which is to their credit.\n\nWhat is actually new: the modular C++ implementation with HOPPET/APPLgrid/MCFM integration, and the first attempt to constrain nPDFs with LHC pPb W/Z data inside the nCTEQ framework. They checked that the APPLgrids are sufficiently PDF-independent, which is a necessary validation. The chi-square improvement from 992/816 to 828/816 is not surprising — you're fitting something and it's better than not fitting — but the fact that the W/Z data pull toward a larger strange sea is worth noting.\n\nWhere it's soft: the fit uses only 12 parameters, with 9 for g, uV, dV, ubar+dbar, compared with 16 in nCTEQ15 for those same flavors. If the data want to move in those freed directions, the fit can compensate through the strange distribution, which is the least constrained. That's a direct confound for the 40% claim, and the paper doesn't test it. The normalization story is also wobbly: with a 1 sigma shift the effect halves, and they allow a few data sets to move up to 3 sigma. Without a proper treatment of the likelihood including normalization penalties, the quoted improvement is not evidence for a physical strange enhancement. There is also no uncertainty band on the 40% shift.\n\nI'd say the right reading is: new tool, promising pipeline, one preliminary observation that needs a full 16-parameter fit and honest treatment of normalization before it means anything. The paper is candid about its limits, and the authors explicitly ask whether the fit is exploiting s(x). That kind of self-assessment deserves a fair referee.\n\nRecommendation: send it to review as a proceedings-style contribution, but the referee should hold the authors to a clear statement that the strange shift is not a determination until they run the full basis.","headline":"A candid progress report whose headline strange shift is likely a byproduct of a restricted parameterization; the framework work is solid and worth a look.","tokens_in":6401,"tokens_out":2729,"would_cite":false,"duration_ms":27024,"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":"Adding LHC proton-lead W/Z data to a nuclear PDF fit raises the lead strange quark distribution by about 40 percent, and the authors ask whether this shift is physical or a fitting artifact.","keywords":["nuclear parton distribution functions","strange quark PDF","W/Z boson production","proton-lead collisions","global QCD fit","nCTEQ","NLO QCD","LHC heavy-ion data"],"falsifier":"Perform the same restricted fit with the three W/Z data sets that need roughly 3σ normalization shifts either removed or forced to zero shift; if the strange PDF returns to its nCTEQ15 value and the remaining W/Z $\\chi^2$ stays far above 1, the 40% increase is a normalization artifact, not a physical signal.","tokens_in":5342,"feed_emoji":"⚛️","tokens_out":9061,"duration_ms":72386,"temperature":0.7,"pith_summary":"The paper claims that the long-standing puzzle of the strange quark parton distribution can be attacked with LHC vector-boson data from proton-lead collisions, and that the new data materially change the extracted nuclear strange PDF. Using a new C++ implementation of the nCTEQ global fit, the authors include NLO W±/Z production data in a preliminary fit alongside the usual DIS and Drell-Yan data. With no normalization shifts allowed, the fit improves the overall $\\chi^2$ per degree of freedom from 992/816 to 828/816 and pulls the lead strange PDF up by about 40% at the $x$ values probed by W/Z production. When normalization shifts of up to 1σ are allowed, that strange-quark increase is roughly halved. The authors state the question directly: are the data telling us the strange sea is larger, or is the fit exploiting the least constrained flavor?","feed_headline":"Strange quark sea in lead jumps 40% with LHC data","feed_subtitle":"Proton-lead W/Z data lift the strange PDF; the authors ask if the shift is real or a fitting artifact.","key_machinery":"The central object is nCTEQ++, a modular C++ rewrite of the nCTEQ nuclear-PDF global-fitting code. It integrates HOPPET for DGLAP evolution, APPLgrid for precomputed NLO coefficient grids, and MCFM for the W/Z matrix elements, and it exports fits in YAML for analysis. The key technical step is validation that the MCFM-generated ApplGrids are sufficiently PDF-independent that proton and nuclear PDF grids can be interchanged inside the fitting loop. The fit described here uses a deliberately limited parameter set—12 parameters instead of nCTEQ15’s 16—so that the strange PDF, normally fixed, is free to move; this restricted setup is what lets the authors attribute changes in $s(x)$ to the new LHC data.","core_discovery":"On its own terms, the paper's central result is that LHC pPb W±/Z data are a new, independent handle on flavor separation in nuclear parton distributions, and specifically that they prefer an enhanced strange PDF in lead. In a restricted fit that frees only 12 parameters—three for $s+\\bar{s}$ and nine for the other fitted flavors—the NLO W/Z data pull $s(x)$ upward by roughly 40% in the $x$ region relevant to vector-boson production, while improving the fit to the W/Z data sets. The same exercise with a 1σ normalization shift cuts that increase roughly in half, which the authors take as a sign that the size of the strange-sea enhancement is entangled with the treatment of data normalization. The paper deliberately stops short of claiming the enhancement is physical; its stated question is whether the shift is dictated by nature or is the fit exploiting $s(x)$ because it is one of the least constrained flavors.","pith_inferences":["If the 40% strange enhancement survives a full 16-parameter fit and the inclusion of ALICE, LHCb, and the second CMS run, it would imply that nuclear shadowing at small $x$ is weaker than nCTEQ15 assumes; the paper does not make this claim, but its own rapidity comparison points in that direction.","A direct cross-check would be to fit neutrino-DIS dimuon data and LHC pPb W/Z data together; if the strange PDF preferred by the two data sets disagree, the tension would locate the problem in nuclear corrections rather than in the strange flavor itself.","The near-3σ normalization shifts on a few data sets could be absorbing missing higher-order QCD corrections or underestimated theory uncertainties; under that reading the physical strange enhancement would be closer to the 1σ-shift result, roughly 20%."],"forward_implications":["If the result holds, LHC proton-lead W/Z measurements become a usable constraint on the strange quark PDF in nuclei, complementing fixed-target neutrino-dimuon data.","The extracted strange PDF depends strongly on the assumed normalization uncertainties: without shifts it rises about 40%, with 1σ shifts about half that, so future fits must quote the normalization treatment alongside the PDF.","Including heavy-ion vector-boson data can improve proton PDFs by reducing the reliance on nuclear corrections applied to heavy-target DIS data.","With normalization shifts of up to about 3σ for a few data sets, a $\\chi^2$/dof near 1 is achievable for all W/Z data, meaning the data are not mutually incompatible once normalizations are free."],"supporting_citations":[{"why":"Supplies the baseline nCTEQ15 global analysis, data set, and fitting framework that the LHC W/Z data are added to.","marker":"[1]"},{"why":"Provides the earlier pPb and PbPb vector-boson predictions and the data-set IDs used in the present fit.","marker":"[11]"},{"why":"Establishes the strange-quark puzzle and the connection between strange PDFs and Drell-Yan/W/Z production.","marker":"[4]"},{"why":"Documents the tension between LHC pp W/Z measurements in the strange sea, the proton-side analogue this paper extends to nuclei.","marker":"[5]"},{"why":"Supplies the APPLgrid technology that lets NLO W/Z cross sections be precomputed and used inside the PDF fit.","marker":"[9]"},{"why":"Provides the MCFM NLO matrix elements from which the W/Z grids are generated.","marker":"[10]"},{"why":"Provides HOPPET, the DGLAP evolution code integrated into nCTEQ++.","marker":"[8]"},{"why":"Earlier analysis of PDF nuclear corrections in charged and neutral current processes, used to interpret the small-x shadowing behavior.","marker":"[12]"},{"why":"Gives the neutrino-nucleus DIS nuclear-correction analysis that the LHC W/Z results are compared against on small-x shadowing and flavor separation.","marker":"[13]"}],"fun_headline_variants":["LHC W/Z data sharpen strange PDF in heavy nuclei","New LHC data boost strange sea in lead, but cautiously","Heavy-ion W/Z data offer new probe of strange quark","pPb W/Z data: strange sea up 40% but fit-dependent","Heavy-ion collisions: new handle on nuclear strange PDF"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The load-bearing assumption is that the normalization shifts allowed for a few LHC data sets—up to about 3σ—are genuine experimental uncertainties rather than a way to absorb theoretical or nuclear-correction errors.","fun_headline_variants_meta":{"raw":{"variants":["LHC W/Z data sharpen strange PDF in heavy nuclei","New LHC data boost strange sea in lead, but cautiously","Heavy-ion W/Z data offer new probe of strange quark","pPb W/Z data: strange sea up 40% but fit-dependent","Heavy-ion collisions: new handle on nuclear strange PDF"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000291,"raw_usage":{"total_tokens":1674,"prompt_tokens":891,"completion_tokens":783,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":507,"completion_tokens_details":{"reasoning_tokens":696}},"tokens_in":507,"tokens_out":783,"duration_ms":6487,"temperature":1.0,"reasoning_tokens":696,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-14T05:52:17.653112+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Perform the same restricted fit with the three W/Z data sets that need roughly 3σ normalization shifts either removed or forced to zero shift; if the strange PDF returns to its nCTEQ15 value and the remaining W/Z $\\chi^2$ stays far above 1, the 40% increase is a normalization artifact, not a physical signal.","supporting_citations":[],"review_version":1}