Pith. sign in

REVIEW 3 major objections 4 minor 14 references

Towards N$^3$LO PDFs and their implications for Higgs production cross-sections

T0 review · 3 major / 4 minor · reviewed 2026-08-15 · deepseek-v4-flash

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗ Review JSON

Pith's one-line read The newest four-loop inputs leave N3LO Higgs cross-sections within a half percent, so the published combined PDF set remains reliable.

desk verdict A short, honest proceedings update from the author of the original combination papers; the central-value stability claim holds up, but the 'conservative uncertainty' inference is not demonstrated. read the letter →

arxiv 2505.10369 v1 pith:ML5TNM6F submitted 2025-05-15 hep-ph

Felix Hekhorn This is my paper · ORCID
classification hep-ph
keywords partondistributionfunctionsapproximateN3LOfour-loopsplittingheavy-quarkoperatormatrixelementsHiggsgluonfusionvector-bosonPDFcombinationLHCphenomenology
verification ladder T0 review T1 audit T2 compute T3 formal

The pith

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The reading

This paper asks whether the newest perturbative inputs—the latest four-loop splitting functions and heavy-quark operator matrix elements—change the recently published combination of approximate $N^3$LO (aN$^3$LO) parton distributions enough to matter for LHC Higgs physics. The author applies these inputs to revised versions of both the MSHT20 and NNPDF4.0 aN$^3$LO fits and compares the results with the original MSHT20$\times$NNPDF4.0 combination. The central PDFs stay within the published 68% confidence bands, and the Higgs cross-section shifts are about 0.5% for gluon fusion and 0.1% for vector-boson fusion. The paper concludes that the original combined set is reliable, that it can be used for $N^3$LO phenomenology at the LHC or the EIC, and that its quoted uncertainty may be treated as conservative. The study is an extension built on revised fits that are reported as forthcoming, not on new global fits performed here.

What carries the argument

The argument runs through the standard factorization formula $F(Q^2)=C(Q^2)\,E(Q^2\leftarrow m_b^2)\,A(m_b^2)\,E(m_b^2\leftarrow Q_0^2)\,f(Q_0^2)$, where $E$ is the DGLAP evolution operator built from the splitting functions, $A$ is the heavy-quark operator matrix element connecting flavour schemes, $C$ is the partonic coefficient function, and $f$ is the fitted PDF at the starting scale. The update replaces $E$ with the newest four-loop splitting functions and $A$ with the latest single-mass three-loop OMEs, and makes MSHT20 adopt the same OMEs as NNPDF4.0 so the two fits differ only in unknown $N^3$LO effects. The combination method is a simple replica sum: 100 replicas from each group's PDF set are drawn and added as a distribution. The paper reads off the consequences in ratio plots and in $N^3$LO matrix-element Higgs cross-sections with PDF and missing-higher-order uncertainties included.

What would settle it

When the revised fits are released, recompute the gluon-fusion and vector-boson-fusion Higgs cross-sections at 13.6 TeV from the final PDF grids. If the gluon-fusion shift from the original combination exceeds about 0.5% for either group, or if the updated PDFs fall outside the published 68% confidence-level bands rather than inside them, the claim that the combination is reliable and its uncertainty conservative would be refuted.

Watch

Extended reading notes

Core claim

The paper's central claim is that the combined approximate-$N^3$LO PDF set MSHT20$\times$NNPDF4.0 is robust against the most recent progress in $N^3$LO calculations. When the two underlying fits are updated to share the same newly known ingredients—the four-loop splitting functions and the latest heavy-quark operator matrix elements—their central values move by less than the existing PDF uncertainty. In Higgs production at $\sqrt{s}=13.6$ TeV, the gluon-fusion cross-section shifts downward by about 0.5% for NNPDF4.0 and upward by slightly less than 0.5% for MSHT20, while vector-boson fusion moves by about 0.1% for both. Because the groups move slightly toward each other and the uncertainties on the new perturbative inputs are reduced, the author concludes that the published combination's uncertainty is conservative, and that the combination can be used to study $N^3$LO PDF effects at the LHC or EIC.

Load-bearing premise

The paper's conclusion depends on the revised MSHT20 and NNPDF4.0 fits, cited as appearing soon, having actually been produced with the stated common splitting functions and heavy-quark matrix elements and with sound statistics; the paper analyzes their plotted outputs but does not reproduce or verify those fits itself.

Editorial extensions

If this is right

  • LHC Higgs predictions made with the published MSHT20xNNPDF40_an3lo combination remain valid: after adding the newest splitting functions and OMEs, gluon-fusion changes by less than about 0.5% and vector-boson fusion by about 0.1%.
  • The published combined set and its uncertainty can be used for approximate-$N^3$LO phenomenology at the LHC and EIC, with the current uncertainty band as a conservative envelope.
  • With common known $N^3$LO inputs, the residual spread between the MSHT20 and NNPDF4.0 variants reflects only the treatment of still-unknown $N^3$LO effects, such as massive DIS coefficient functions and hadronic $K$-factors.
  • The dominant future reduction in PDF-induced uncertainty for inclusive Higgs production will have to come from the partonic matrix elements $C$, the one ingredient not yet available at $N^3$LO.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

  • Editorial inference: because the two updated fits shift in opposite directions and toward each other, a future recombination of the revised fits could yield a smaller central spread and a narrower uncertainty than the original combination, not merely the same conservative band.
  • Editorial inference: the few-per-mille size of the inclusive shifts suggests that residual unknown-$N^3$LO PDF freedom will show up more in differential distributions, where the gluon is probed at different $x$ values, than in inclusive cross-sections.
  • Editorial inference: the same update-and-combine procedure could be applied to small-$x$ observables relevant for the EIC; at very small $x$ the gluon uncertainty is larger, so the conservative-uncertainty conclusion may not transfer automatically.
Share X Bluesky LinkedIn Reddit HN

Signed reviews

No signed human review yet.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Referee Report

3 major / 4 minor

Summary. The paper reviews the recently published combined aN3LO PDF sets MSHT20xNNPDF40_an3lo and the QED variant, and investigates whether the most recent four-loop splitting functions (FHMRUVV) and new heavy-quark operator matrix elements (OMEs) alter the underlying MSHT20 and NNPDF4.0 aN3LO fits. The author reports that adopting these common inputs shifts the gluon and singlet PDFs at Q = 100 GeV within the 68% confidence bands, that Higgs gluon-fusion cross-sections move by about 0.5% (down for NNPDF, up for MSHT) and vector-boson fusion by about 0.1%, and concludes that the original combination is reliable and can be used for N3LO PDF phenomenology at the LHC and EIC, with the original uncertainty estimate possibly conservative.

Significance. If the stated conclusions are fully supported, the paper provides a useful and timely cross-check of the aN3LO PDF combination, showing that the latest perturbative ingredients do not materially change central predictions for two key LHC Higgs channels. The manuscript is honest about its preliminary character and clearly attributes the new splitting functions and OMEs to independent calculations, which is a strength. However, the significance is limited by the absence of publicly available revised fits, the lack of numerical tables, and the fact that the 'conservative uncertainty' claim is asserted rather than demonstrated. The paper's main value is as a short phenomenological update rather than as a self-contained derivation.

major comments (3)
  1. [Section 2, Ref. 12] The central quantitative results depend entirely on revised MSHT20 and NNPDF4.0 aN3LO fits that are only cited as 'to appear' (Ref. 12). The paper provides no PDF grids, no chi-square or parameter-shift information, and no validation that these revised fits indeed use the stated common inputs (FHMRUVV splitting functions and, for MSHT, the NNPDF OMEs). Consequently, the reliability conclusion and the quoted cross-section shifts cannot be independently checked. This is load-bearing for the paper's main claim and needs to be addressed, either by making the revised fits public or by clearly restricting the conclusions to the central-value trends shown in the figures.
  2. [Section 3] The statement that 'the theoretical uncertainties will automatically be improved by the reduction in uncertainty in N3LO splitting functions and (in the case of MSHT20) OMEs' is an assumption, not a demonstrated result. Improved perturbative inputs can shift fitted PDFs and can in principle increase data-theory tension, and the scale-variation/MHOU prescription need not respond monotonically. The paper provides no comparison of the MHOU bands between the original and revised PDF sets, so the conclusion that 'the uncertainty in our official combination may be taken as conservative' is unsupported. Please either provide the relevant uncertainty comparison or soften this conclusion to a conjecture.
  3. [Figures 1 and 2] The paper quotes specific central-value shifts ('about 0.5% for ggF', 'of order 0.1% for VBF') but gives no numerical tables and no uncertainty estimates for these shifts. The figures show ratios within 68% bands, which supports the qualitative stability claim, but a quantitative claim about the size of the effect requires numbers with uncertainties, especially because the PDF and MHOU errors are correlated between the original and revised sets. Please provide a small table with central values and total uncertainties for the cross-sections shown in Fig. 2.
minor comments (4)
  1. [Abstract] The phrase 'we demonstrate the impact' is stronger than the body's explicit 'preliminary' framing; consider aligning the abstract with the paper's stated preliminary character.
  2. [Section 1, Eq. (1)] The notation 'between between the four and five flavour scheme' contains a duplicated word; please correct this typographical error.
  3. [Figure 1 caption] The caption says 'the latest OMEs' but the surrounding text indicates the OME update applies to the MSHT20 fit; please specify precisely which fit each panel refers to, and clarify what 'FHMRUVV update' includes for each group.
  4. [Section 3, Fig. 2] The legend labels 'NNPDF40 rev.aN3LO' and 'MSHT20 rev.aN3LO' are not defined in the text; please define 'rev.' explicitly at first use.

Circularity Check

0 steps flagged · score 0.0 of 10

No circularity: cross-sections are computed from externally fitted PDFs and independently calculated perturbative ingredients; no fitted quantity is renamed as a prediction.

full rationale

The paper contains no derivation chain in which a claimed prediction is equivalent to its input by construction. The updated MSHT20 and NNPDF4.0 aN3LO fits are obtained by refitting with the FHMRUVV four-loop splitting functions (Ref. 10) and the new heavy-quark OMEs (Ref. 11), which are computed independently of the Higgs cross-sections shown in Fig. 2. The Higgs cross-sections are then evaluated with fixed-order matrix elements; they are not fed back into any fit, so the propagation from PDFs to cross-sections is a standard convolution, not a self-definitional cycle. The central-value stability check (all shifts within the 68% c.l. PDF uncertainties) is a consistency comparison between the original and revised PDF sets, not a circular validation. The statement that the official combination's uncertainty 'may be taken as conservative' is an assumption based on expected reduction in theory uncertainty from improved splitting functions/OMEs, and the text itself marks the results as preliminary ('defer a detailed investigation to future publications'); this is a support gap rather than circularity. Ref. 12 ('to appear') is invoked for the revised fits, but no load-bearing argument reduces to that citation alone, and the paper does not deploy any uniqueness theorem from the authors' prior work. Accordingly, there is no circular step.

Assumptions & free parameters 0 free parameters · 3 assumptions · 0 invented entities

No new free parameters or invented entities are introduced in this paper. The central claim inherits the fitted parameters of the underlying MSHT20 and NNPDF4.0 aN3LO PDFs from prior work, and relies on standard QCD assumptions plus the paper's own choice to combine replicas by simple averaging. The validity of the cited revised fits (Ref. 12) is an external reliability assumption, flagged in weakest_assumption.

assumptions (3)
  • domain assumption QCD factorization and DGLAP evolution at approximate N3LO
    The whole PDF extraction and cross-section computation relies on factorization and DGLAP, stated in Eq. (1) and the surrounding text.
  • domain assumption Approximate N3LO framework: perturbative ingredients not fully known are approximated
    Both groups use aN3LO approximations; the paper relies on this shared scheme to compare fits and to claim updates only change predictions mildly.
  • ad hoc to paper Simple Monte Carlo replica averaging of the two PDF sets yields a valid combined distribution
    Section 1: 'The combination is performed by drawing 100 replicas respectively from the publicly available PDF sets and summing them plainly.' This is a modeling choice for the new combined set, assumed valid without validation.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Towards N$^3$LO PDFs and their implications for Higgs production cross-sections." pith.science (2026). https://pith.science/paper/ML5TNM6F

@misc{pith2026250510369,
  author       = {Pith},
  title        = {Pith review of: Towards N$^3$LO PDFs and their implications for Higgs production cross-sections},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/ML5TNM6F}},
  note         = {Machine review of arXiv:2505.10369}
}
abstract

We review the recently published PDF sets MSHT20xNNPDF40_an3lo and MSHT20xNNPDF40_an3lo_qed, which have been obtained by combining the approximate N$^3$LO (aN$^3$LO) sets of parton distributions (PDFs) from the MSHT20 and NNPDF4.0 series. We investigate the impact of the most recent results on N$^3$LO splitting functions onto the two PDF sets. Finally, we demonstrate the impact of N$^3$LO PDFs on the computation of Higgs production cross-sections at the LHC.

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

14 extracted references · 2 canonical work pages

  1. [1]

    Cridge et al., Combination of aN ^3 LO PDFs and implications for Higgs production cross-sections at the LHC , http://arxiv.org/abs/2411.05373 arXiv:2411.05373

    MSHT, NNPDF Collaboration, T. Cridge et al., Combination of aN ^3 LO PDFs and implications for Higgs production cross-sections at the LHC , http://arxiv.org/abs/2411.05373 arXiv:2411.05373

    arXiv

  2. [2]

    Bailey, T

    S. Bailey, T. Cridge, L. A. Harland-Lang, A. D. Martin, and R. S. Thorne, Parton distributions from LHC, HERA, Tevatron and fixed target data: MSHT20 PDFs , Eur. Phys. J. C 81 (2021), no. 4 341, [ http://arxiv.org/abs/2012.04684 arXiv:2012.04684 ]

    arXiv 2021

  3. [3]

    NNPDF Collaboration, R. D. Ball et al., The path to proton structure at 1\

    work page

  4. [4]

    McGowan, T

    J. McGowan, T. Cridge, L. A. Harland-Lang, and R. S. Thorne, Approximate N ^ 3 LO parton distribution functions with theoretical uncertainties: MSHT20aN ^3 LO PDFs , Eur. Phys. J. C 83 (2023), no. 3 185, [ http://arxiv.org/abs/2207.04739 arXiv:2207.04739 ]. [Erratum: Eur.Phys.J.C 83, 302 (2023)]

    arXiv 2023

  5. [5]

    NNPDF Collaboration, R. D. Ball et al., The path to N ^3 LO parton distributions , Eur. Phys. J. C 84 (2024), no. 7 659, [ http://arxiv.org/abs/2402.18635 arXiv:2402.18635 ]

    arXiv 2024

  6. [6]

    Cridge, L

    T. Cridge, L. A. Harland-Lang, and R. S. Thorne, Combining QED and approximate N ^3 LO QCD corrections in a global PDF fit: MSHT20qed\_an3lo PDFs , SciPost Phys. 17 (2024), no. 1 026, [ http://arxiv.org/abs/2312.07665 arXiv:2312.07665 ]

    arXiv 2024

  7. [7]

    Barontini, N

    A. Barontini, N. Laurenti, and J. Rojo, NNPDF4.0 aN ^3 LO PDFs with QED corrections , in 31st International Workshop on Deep-Inelastic Scattering and Related Subjects , 6, 2024. http://arxiv.org/abs/2406.01779 arXiv:2406.01779

    arXiv 2024

  8. [8]

    Caola, W

    F. Caola, W. Chen, C. Duhr, X. Liu, B. Mistlberger, F. Petriello, G. Vita, and S. Weinzierl, The Path forward to N ^3 LO , in Snowmass 2021 , 3, 2022. http://arxiv.org/abs/2203.06730 arXiv:2203.06730

    arXiv 2021

Show all 14 references
  1. [9]

    J. A. M. Vermaseren, A. Vogt, and S. Moch, The Third-order QCD corrections to deep-inelastic scattering by photon exchange , Nucl. Phys. B 724 (2005) 3--182, [ http://arxiv.org/abs/hep-ph/0504242 hep-ph/0504242 ]

    2005 arXiv

  2. [10]

    Falcioni, F

    G. Falcioni, F. Herzog, S. Moch, A. Pelloni, and A. Vogt, Four-loop splitting functions in QCD the gluon-gluon case , Phys. Lett. B 860 (2025) 139194, [ http://arxiv.org/abs/2410.08089 arXiv:2410.08089 ]

    2025 arXiv

  3. [11]

    umlein, A. De Freitas, A. von Manteuffel, C. Schneider, and K. Sch\

    J. Ablinger, A. Behring, J. Bl\"umlein, A. De Freitas, A. von Manteuffel, C. Schneider, and K. Sch\"onwald, The non-first-order-factorizable contributions to the three-loop single-mass operator matrix elements AQg(3) and AQg(3) , Phys. Lett. B 854 (2024) 138713, [ http://arxiv...

    2024 arXiv

  4. [12]

    Cridge et al., `` Approximate N3LO PDFs: Updates and Consequences for Phenomenology .'' To appear in DIS2025, 2025

    T. Cridge et al., `` Approximate N3LO PDFs: Updates and Consequences for Phenomenology .'' To appear in DIS2025, 2025

    2025

  5. [13]

    PDF4LHC Working Group Collaboration, R. D. Ball et al., The PDF4LHC21 combination of global PDF fits for the LHC Run III , J. Phys. G 49 (2022), no. 8 080501, [ http://arxiv.org/abs/2203.05506 arXiv:2203.05506 ]

    2022 arXiv

  6. [14]

    write newline

    " write newline "" before.all 'output.state := FUNCTION blank.sep after.quote 'output.state := FUNCTION fin.entry output.state after.quoted.block = 'skip 'add.period if write newline FUNCTION new.block output.state before.all = 'skip output.state after.quote = after.quoted.blo...

Pith tools

Reviewed August 15, 2026 · model on record in the stance chip above.