{"id":"bb37dc5e-c094-4e38-8667-340f723e73c1","arxiv_id":"1908.04983","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":5.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":3,"one_line_summary":"A new NNLO nuclear PDF fit, TUJU19, built on a modified xFitter framework, is presented and shown to be broadly consistent with existing sets.","lead":"The authors present TUJU19, a new set of nuclear parton distribution functions (nPDFs) extracted from deep inelastic scattering data at NLO and NNLO. The fit uses a modified open-source tool xFitter and gives results broadly consistent with existing nPDF sets, with a notable gluon difference at large x at low scale that fades at higher scales.","discovery_kind":"extension","skeptic_critique":{"model":"deepseek-v4-flash","headline":"TUJU19 central-value comparisons lack uncertainty bands, so the claims of agreement and of a confined gluon deviation are not yet quantitatively supported.","rationale":"The reader's verdict CONDITIONAL is appropriate for a proceedings paper that defers quantitative details to a companion publication. My stress-test focuses on the most load-bearing gap in the central claim: the comparison of TUJU19 central values to other sets' error bands without TUJU19's own uncertainty, and the unsupported assertion of higher-scale agreement. This is closely related to the reader's red flag about missing error bars, but the reader's stated weakest assumption was the functional form of Eq. (2.2) and the fixed proton baseline. I consider the missing uncertainty quantification to be more directly load-bearing because, even if the parameterization were perfectly adequate, the claim of consistency or of a confined gluon deviation cannot be quantitatively assessed without TUJU19's uncertainties and a statistical comparison. The proposed concrete test (pull computation with full TUJU19 uncertainties at two scales) would settle whether the central claim survives. Since this concern reinforces the reader's CONDITIONAL verdict rather than overturning it, the verdict remains UNCHANGED.","tokens_in":5741,"tokens_out":5913,"duration_ms":63784,"concrete_test":"Compute the full Hessian or Monte Carlo uncertainty set for TUJU19 at NLO and NNLO, then evaluate the pulls between TUJU19 and nCTEQ15/EPPS16 for the Pb-to-proton ratios R_g, R_dv, and R_dbar at Q^2 = 1.69 GeV^2 and at a higher scale such as Q^2 = 100 GeV^2, using pull_i = (R_i^TUJU19 - R_i^ref) / sqrt(sigma_TUJU19^2 + sigma_ref^2). If all |pull| < 1.5, the claimed deviations are not significant; if the gluon pull exceeds 2 at the higher scale, the claim of higher-scale agreement fails. The same exercise should be repeated in the NNLO comparison once a suitable NNLO reference set is available.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central claim in Section 4—that TUJU19 central PDFs are mostly within the error bands of nCTEQ15 and EPPS16, with only the large-x gluon modification deviating at the initial scale and agreement restored at higher scales—rests on a comparison of TUJU19 central values against the published uncertainty bands of the other sets. No uncertainty estimate for TUJU19 is shown, and the paper explicitly defers error bands to the companion publication [11]. This makes the claim asymmetric and quantitatively underdetermined: a central value can lie inside another set's error band while the difference is still highly significant if TUJU19's own uncertainties are small, or the reverse if TUJU19's uncertainties are large. In addition, the statement that 'agreement is found at higher scales' is not supported by any figure, table, or quantitative criterion in this paper; it is an assertion about an unpublished comparison. The load-bearing assumption is therefore not a specific physics input but the implicit assumption that the displayed central-value comparisons are meaningful evidence of consistency. Without TUJU19's uncertainty bands and a quantitative measure of agreement (e.g., pulls or chi-square), the central claim cannot be evaluated, and the claimed 'confined' nature of the gluon deviation could be an artifact of either the fixed proton baseline in Eq. (2.2) or the restrictive parameterization.","agreement_with_reader":"partial"},"referee_report":{"model":"deepseek-v4-flash","summary":"This proceedings paper presents a new QCD analysis of nuclear parton distribution functions (nPDFs), dubbed TUJU19, performed at both NLO and NNLO using a modified version of the open-source xFitter framework. The authors first construct a free-proton baseline from HERA, BCDMS, and NMC data, then parameterize nuclear modifications through A-dependent coefficients of the form c_k(A) = c_{k,0} + c_{k,1}(1 - A^{-c_{k,2}}), fitting 16 nuclear parameters to neutral-current DIS and charged-current neutrino-nucleus DIS data. The paper shows representative data comparisons and compares TUJU19 central values with nCTEQ15 and EPPS16 at NLO. The central qualitative claim is that TUJU19 is consistent with existing nPDF sets except for a large-x gluon nuclear modification at the initial scale, with agreement restored at higher scales.","tokens_in":6041,"tokens_out":2389,"duration_ms":26745,"significance":"If fully validated, TUJU19 would provide a useful new nPDF set, notable for being based on an open-source fitting framework and for treating the proton baseline consistently with the nuclear fit. The explicit A-dependent parameterization and the inclusion of neutrino-nucleus charged-current data are positive aspects. The paper also demonstrates that NLO and NNLO fits give comparable descriptions of the included data, which is a useful consistency check. However, the quantitative support for these claims is currently missing: no uncertainty bands for TUJU19, no chi-square values, and no plot or table for the claimed higher-scale gluon agreement. The paper is therefore best viewed as a preliminary proceedings contribution whose central comparisons need further support to be fully assessed.","major_comments":[{"comment":"The central claim that TUJU19 central PDFs are 'mostly within the error bands' of nCTEQ15 and EPPS16 is not quantitatively supported, because TUJU19 is shown only as central values and the paper explicitly defers its own uncertainty bands to the companion publication [11]. A central value lying inside another set's error band is insufficient to establish consistency unless TUJU19's own uncertainties are also known; the significance of any difference depends on both sets' errors. Please either include TUJU19 uncertainty bands or reframe the comparison as a qualitative central-value check and clearly state that no quantitative consistency claim can yet be made.","section":"Section 4, Figure 4"},{"comment":"The statement that 'the agreement of the calculated quantities with the measurements is very good at NLO and NNLO' is based only on a selected representative subset, with no chi-square values or pulls per data set. Since these are fitted data, good agreement with a few selected points does not quantify the fit quality. Please provide a table of chi-square or equivalent goodness-of-fit measures for the full data set, or explicitly restrict the claim to the displayed representative subset.","section":"Section 4, Figures 2 and 3"},{"comment":"The sentence 'agreement is found at higher scales' is not accompanied by any figure, table, or quantitative criterion in this paper. As written, it is an assertion about an unpublished comparison. Either show the higher-scale comparison or remove this statement or mark it as a forthcoming result, so that the reader can distinguish between demonstrated and deferred claims.","section":"Section 4, last paragraph"}],"minor_comments":[{"comment":"The name of the fitting tool is written inconsistently as 'XFITTER' in the abstract and 'xFitter' in the main text; please standardize capitalization.","section":"Abstract and text"},{"comment":"The caption contains a typo: 'brakets' should be 'brackets'.","section":"Figure 1 caption"},{"comment":"The phrase 'the variance in the amplitude for the sea quark distributions is quite large' would be clearer as 'the variation in the amplitude'.","section":"Section 4, Figure 1"},{"comment":"The parameterization ansatz is written with the same symbols c0...c4 for both the proton and nuclear PDFs; please clarify that the nuclear coefficients are the A-dependent ones from Eq. (2.2), while the proton coefficients are the fixed c_{k,0} values, to avoid notational confusion.","section":"Section 2, Eq. (2.1)"},{"comment":"The comparison to nCTEQ15 and EPPS16 is shown only at NLO; the text does not comment on whether an analogous NNLO comparison was attempted. A sentence stating why only NLO is shown, or what the NNLO trend is, would help.","section":"Section 4, Figure 4"}],"recommendation":"major_revision","confidential_remarks":"This is a proceedings contribution whose main numerical claims are explicitly deferred to a companion paper. The core issue is not the physics assumption but the gap between the claims made in Section 4 and the evidence shown. I would suggest that the authors either add the missing uncertainty/chi-square material or explicitly mark the comparison claims as preliminary. The paper has a reasonable place as a proceedings record if the claims are appropriately scoped."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Short version: this is a solid proceedings write-up of a preliminary nPDF extraction (TUJU19) at NLO and NNLO. What's actually new: the authors build their own proton baseline, then fit A-dependent nuclear parameters with a modified xFitter, including neutral-current DIS and neutrino charged-current DIS in one fit, and treat deuteron as a nucleus. The open-source framework detail is the most useful part. The plots show plausible agreement with the fitted data, and the comparison to nCTEQ15 and EPPS16 is a useful sanity check.\n\nThe soft spots are exactly the ones the reader flags. There are no uncertainty bands for TUJU19 anywhere, no chi-square values, and only a selected subset of data is shown. The sentence in Sec. 4, \"the central PDFs are mostly within the error bands of the other sets,\" therefore cannot be evaluated: comparing a central value to someone else's error band is not a quantitative statement about consistency. The follow-up claim that agreement is restored at higher scales is not backed by any figure in this paper. For a proceedings that might be tolerable if the wording were suitably provisional; here it reads as stronger than the evidence.\n\nThe bigger structural issue: the only source of quantitative support is the companion paper [11], which appears as arXiv:1908.03355. That is fine for a proceedings, but it means the standalone value of this paper is limited. On the physics side, the fixed proton baseline ck,0 and the restrictive form of Eq. (2.2) are standard choices, not red flags, but they do mean the nuclear modifications inherit any bias in the proton fit. That is worth a sentence in the conclusions.\n\nI don't think there is a fatal flaw. The methodology is coherent, and the comparison plots are consistent with what an independent nPDF fit should look like. The paper's claims just outrun its presented evidence. For a full journal article I would want the uncertainty analysis and chi-square tables in the same paper; for a proceedings, the authors should soften the comparison claims and clearly mark the results as preliminary.\n\nWho is it for? Nuclear PDF practitioners and people building on xFitter for nuclear fits. It deserves a serious referee: the framework is real, the modifications are concrete, and the companion paper is likely to matter. My recommendation: engage with it, but require the overstated claims be either quantified or explicitly deferred.","headline":"A solid proceedings write-up of a preliminary nPDF extraction whose quantitative claims outrun the presented evidence: no uncertainty bands, no chi-square values, and the key consistency statement is not evaluable as written.","tokens_in":6501,"tokens_out":2130,"would_cite":false,"duration_ms":23034,"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":"New nuclear PDFs match earlier extractions except one gluon shift.","keywords":["nuclear parton distribution functions","NNLO QCD analysis","deep inelastic scattering","neutrino-nucleus scattering","DGLAP evolution","isospin symmetry","nPDF parameterization","TUJU19"],"falsifier":"A future low-scale measurement of the nuclear gluon ratio $R_g = xg^{p/Pb}/xg^p$ at $x \\gtrsim 0.2$ and $Q^2 \\approx Q_0^2$ that follows the earlier nPDF sets rather than the TUJU19 large-$x$ enhancement, together with a re-fit showing that the discrepancy cannot be removed within uncertainties, would settle the claim.","tokens_in":5544,"feed_emoji":"⚛️","tokens_out":6030,"duration_ms":59583,"temperature":0.7,"pith_summary":"This paper presents a new set of nuclear parton distribution functions (nPDFs), named TUJU19, determined at next-to-leading and next-to-next-to-leading order from deep-inelastic scattering data on several nuclear targets. The authors build their own free-proton baseline from the same parameterization, extend it to bound nucleons with an A-dependent coefficient form, and modify an open-source fitting tool so it can handle data published as cross-section ratios and isoscalar-corrected structure functions. They fit 16 nuclear parameters to neutral-current DIS and charged-current neutrino-nucleus data, including deuteron as a nucleus with nonzero nuclear effects. The central claim is that the resulting nuclear PDFs agree with earlier sets within uncertainties, with the single exception of the gluon nuclear modification at large $x$ at the initial scale $Q_0^2 = 1.69\\,\\mathrm{GeV}^2$, which converges to agreement at higher scales.","feed_headline":"New nuclear PDFs match earlier fits except one gluon shift","feed_subtitle":"TUJU19 provides NLO and NNLO nuclear parton distributions from a self-consistent proton baseline and neutrino-nucleus data.","key_machinery":"The load-bearing object is the A-dependent parameterization $$x $f_i^{{p/A}}$(x,$Q_0^{2}$)=c_0(A)\\,$x^{{c_1(A)}}$(1-x)^{c_2(A)}\\left(1+c_3(A)x+c_4(A)$x^{2}$\\right)$$ with $$c_k(A)=c_{k,0}+c_{k,1}\\left(1-$A^{{-c_{k,2}}$}\\right),$$ where the $c_{k,0}$ are fixed from the fitted proton PDFs and only $c_{k,1}$ and $c_{k,2}$ are free nuclear parameters. Bound-neutron PDFs follow by isospin symmetry, and the nuclear PDF is constructed as $(Z f^{p/A}+(A-Z) f^{n/A})/A$. This form reduces to the free proton at $A=1$. Around it, the analysis uses DGLAP evolution at NLO or NNLO with the corresponding splitting functions and structure-function coefficients, and a modified version of an open-source fitting tool that compares theory to absolute cross sections, cross-section ratios, and isoscalar-corrected data.","core_discovery":"The central discovery is that a self-consistent simultaneous analysis of neutral-current and charged-current neutrino-nucleus DIS data can determine nuclear PDFs at NLO and NNLO, and that these PDFs agree with existing extractions in most regions. The only notable difference is in the gluon sector: at the initial scale, the ratio $R_g^{p/Pb}$ rises at large $x$ relative to earlier analyses, but after DGLAP evolution to higher $Q^2$ the difference disappears. The paper therefore claims TUJU19 is a valid alternative to existing nPDF sets, with the advantage of an internally fitted proton baseline and a common fit of neutrino-nucleus charged-current data together with neutral-current data.","pith_inferences":["A natural next test would be to add Drell-Yan data, which the authors plan; if the large-$x$ gluon difference persists there, it would indicate genuine nuclear dynamics rather than a parameterization artifact.","Because the proton coefficients are fixed beforehand, the framework could be re-used to assess how sensitive the extracted nuclear PDFs are to the choice of proton baseline.","Comparing TUJU19 to a fit where deuterium is treated as a free proton would quantify how much of the extracted nuclear effects are driven by the deuteron assumption.","If a future electron-ion collider measures nuclear structure functions in the region where TUJU19 and earlier sets differ, the A-dependent form's flexibility can be tested directly against data."],"forward_implications":["The TUJU19 set provides NLO and NNLO nuclear PDFs that can be used for predictions of heavy-ion and future electron-ion collider observables.","Charged-current neutrino-nucleus DIS data can be included in a common fit with neutral-current DIS data, reproducing both data types well at NLO and NNLO.","Treating deuteron as a nucleus with nonzero nuclear effects gives a definite nuclear correction for deuterium-based measurements.","The large-$x$ gluon difference from earlier extractions is confined to the initial scale and disappears after evolution, so the gluon is consistent with existing sets at higher scales.","The modified open-source fitting tool can handle cross-section ratios and isoscalar corrections, enabling future nuclear PDF analyses of additional process types."],"supporting_citations":[{"why":"Supplies the A-dependent functional form for the fit coefficients used in Eq. (2.2).","marker":"[7]"},{"why":"Supplies the proton PDF parameterization ansatz that the nuclear coefficients extend.","marker":"[6]"},{"why":"Provides the NNLO splitting functions that drive the DGLAP evolution in the analysis.","marker":"[3]"},{"why":"Completes the NNLO splitting-function input for the evolution.","marker":"[4]"},{"why":"Provides the NNLO structure-function coefficients used to compute DIS cross sections.","marker":"[5]"},{"why":"Provides the open-source fitting tool that was modified for nuclear PDF analysis.","marker":"[8]"},{"why":"Defines the neutrino-nucleus charged-current cross-section treatment and an earlier nPDF set for comparison.","marker":"[10]"},{"why":"Provides one of the existing NLO nuclear PDF sets used for comparison in the results.","marker":"[12]"},{"why":"Provides a recent NNLO nuclear PDF analysis compared against in the discussion.","marker":"[14]"},{"why":"Gives the forthcoming full publication containing the error analysis and dataset details.","marker":"[11]"}],"fun_headline_variants":["Nuclear PDFs at NNLO: match old, one gluon shift fades","TUJU19 nPDFs: consistent, only large-x gluon rises at Q0","New NNLO nuclear fits agree, but gluon bump evens out","QCD nuclear analysis: same PDFs, one gluon difference gone","Nuclear PDFs: NLO+NNLO fit, only gluon deviates initially"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The nuclear modifications are assumed to be completely described by the A-dependent parameterization with proton coefficients fixed in advance, so if that functional form is too restrictive or the proton baseline is biased, the extracted nuclear PDFs shift.","fun_headline_variants_meta":{"raw":{"variants":["Nuclear PDFs at NNLO: match old, one gluon shift fades","TUJU19 nPDFs: consistent, only large-x gluon rises at Q0","New NNLO nuclear fits agree, but gluon bump evens out","QCD nuclear analysis: same PDFs, one gluon difference gone","Nuclear PDFs: NLO+NNLO fit, only gluon deviates initially"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000359,"raw_usage":{"total_tokens":1873,"prompt_tokens":803,"completion_tokens":1070,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":419,"completion_tokens_details":{"reasoning_tokens":965}},"tokens_in":419,"tokens_out":1070,"duration_ms":9444,"temperature":1.0,"reasoning_tokens":965,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-14T13:26:30.235112+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"A future low-scale measurement of the nuclear gluon ratio $R_g = xg^{p/Pb}/xg^p$ at $x \\gtrsim 0.2$ and $Q^2 \\approx Q_0^2$ that follows the earlier nPDF sets rather than the TUJU19 large-$x$ enhancement, together with a re-fit showing that the discrepancy cannot be removed within uncertainties, would settle the claim.","supporting_citations":[{"cited_title":"Kovarik et al., Phys.Rev","cited_arxiv_id":null,"evidence_quote":"Supplies the A-dependent functional form for the fit coefficients used in Eq. (2.2)."},{"cited_title":"Abramowicz et al., Eur.Phys.J","cited_arxiv_id":null,"evidence_quote":"Supplies the proton PDF parameterization ansatz that the nuclear coefficients extend."},{"cited_title":"Moch, J.A.M","cited_arxiv_id":null,"evidence_quote":"Provides the NNLO splitting functions that drive the DGLAP evolution in the analysis."},{"cited_title":"Moch, J.A.M","cited_arxiv_id":null,"evidence_quote":"Completes the NNLO splitting-function input for the evolution."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Provides the NNLO structure-function coefficients used to compute DIS cross sections."},{"cited_title":"Zenaiev for the collaboration), PoS DIS2016 (2016) 033","cited_arxiv_id":null,"evidence_quote":"Provides the open-source fitting tool that was modified for nuclear PDF analysis."},{"cited_title":"de Florian, R","cited_arxiv_id":null,"evidence_quote":"Defines the neutrino-nucleus charged-current cross-section treatment and an earlier nPDF set for comparison."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Provides one of the existing NLO nuclear PDF sets used for comparison in the results."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Provides a recent NNLO nuclear PDF analysis compared against in the discussion."},{"cited_title":"Open-source QCD analysis of nuclear parton distribution functions at NLO and NNLO","cited_arxiv_id":"1908.03355","evidence_quote":"Gives the forthcoming full publication containing the error analysis and dataset details."}],"review_version":1}