arxiv: 2604.14133 · v1 · submitted 2026-04-15 · ✦ hep-ph · hep-ex· nucl-ex· nucl-th

Recognition: unknown

AI-assisted modeling and Bayesian inference of unpolarized quark transverse momentum distributions from Drell-Yan data

Zhong-Bo Kang , Luke Sellers , Congyue Zhang , Curtis Zhou

Authors on Pith no claims yet

Pith reviewed 2026-05-10 12:55 UTC · model grok-4.3

classification ✦ hep-ph hep-exnucl-exnucl-th

keywords TMD PDFsDrell-YanBayesian inferenceMachine learningCollins-Soper kernelQuark distributionsGlobal fitNonperturbative QCD

0 comments

The pith

Bayesian inference with AI-selected models and machine-learning surrogates extracts unpolarized quark TMD PDFs from global Drell-Yan data at N3LO+N4LL accuracy.

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

The paper develops a Bayesian framework that uses artificial intelligence to select functional forms for nonperturbative TMD contributions and the Collins-Soper kernel, then builds a machine-learning emulator to accelerate cross-section evaluations. This setup permits efficient Markov Chain Monte Carlo sampling over a global set of Drell-Yan measurements from fixed-target, RHIC, and LHC experiments. A reader would care because TMD PDFs encode the transverse motion of quarks inside protons, and the method supplies these distributions together with uncertainty estimates at high perturbative order.

Core claim

The authors present an extraction of unpolarized quark transverse-momentum-dependent parton distribution functions from Drell-Yan data within a Bayesian inference framework that incorporates artificial intelligence at multiple stages. The analysis reaches N3LO in perturbative QCD combined with N4LL resummation. An AI-driven iterative procedure ranks candidate functional forms for the nonperturbative parts of the TMD PDFs and the Collins-Soper kernel using chi-squared fits and physics constraints. A machine-learning emulator is trained to serve as a surrogate model for the TMD cross sections, replacing repeated expensive evaluations and enabling scalable affine-invariant MCMC sampling. The 결과

What carries the argument

Machine-learning emulator trained as a surrogate model for TMD cross sections, which replaces computationally expensive repeated evaluations and permits efficient MCMC sampling inside the Bayesian global fit.

If this is right

The framework performs a global analysis of Drell-Yan data from fixed-target, RHIC, and LHC experiments at N3LO + N4LL accuracy.
TMD PDFs are obtained together with quantified uncertainties that can be compared to results from the replica method.
Differences in the resulting uncertainty estimates between the Bayesian and replica approaches are highlighted.
The AI procedure ranks functional forms for nonperturbative TMD pieces and the Collins-Soper kernel using chi-squared and physics constraints.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

The surrogate approach could be applied to other processes such as semi-inclusive deep-inelastic scattering to obtain more complete sets of TMD distributions.
If the AI ranking criteria miss important physical constraints, the selected functional forms might still underfit or overfit in ways not captured by the current uncertainty bands.
Scalable sampling opens the possibility of including higher-order corrections or additional data sets without prohibitive computational cost.

Load-bearing premise

The machine-learning surrogate reproduces the true TMD cross sections accurately across the sampled parameter space, and the AI-ranked functional forms for nonperturbative contributions and the Collins-Soper kernel are flexible and unbiased enough to fit the data without systematic artifacts.

What would settle it

Direct numerical comparison showing that the trained emulator deviates by more than the reported uncertainties from exact TMD cross-section calculations at parameter points visited by the MCMC chain would invalidate the extracted distributions and their uncertainties.

read the original abstract

We present an extraction of unpolarized quark transverse-momentum-dependent parton distribution functions (TMD PDFs) from Drell-Yan data within a Bayesian inference framework, incorporating artificial intelligence at multiple stages of the analysis. Our analysis is performed at ${\rm N^3LO}$ in perturbative QCD combined with ${\rm N^4LL}$ resummation accuracy. We first employ an AI-driven iterative procedure to explore and rank candidate functional forms for the nonperturbative contributions to TMD PDFs at the initial scale, as well as for the Collins-Soper evolution kernel, using $\chi^2$ fits and physics constraints. To enable efficient Bayesian inference, we construct a surrogate model for TMD cross sections by training a machine-learning emulator over the parameter space, replacing computationally expensive repeated evaluations and allowing scalable sampling with an affine-invariant Markov Chain Monte Carlo (MCMC) ensemble. Using this framework, we perform a global analysis of Drell-Yan data from fixed-target, RHIC, and LHC experiments and extract TMD PDFs with quantified uncertainties. We compare the results with those obtained using the replica method and highlight differences in the resulting uncertainty estimates.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Desk Editor's Note private letter to a colleague

The paper uses AI to rank functional forms for nonperturbative TMD contributions and the Collins-Soper kernel, then trains a surrogate emulator to run full Bayesian MCMC on Drell-Yan data at N3LO+N4LL.

read the letter

The main thing here is that the authors automate the choice of functional forms for the nonperturbative TMD PDFs at the initial scale and for the Collins-Soper kernel with an AI-driven iterative search based on chi-squared and physics constraints. They then train a machine-learning emulator for the TMD cross sections so they can afford proper affine-invariant MCMC sampling in a Bayesian framework instead of relying only on replicas or point fits. They apply this to a global Drell-Yan dataset from fixed-target, RHIC, and LHC experiments and extract the unpolarized quark TMDs with uncertainty bands, noting differences from the replica approach. This combination of automated ansatz ranking plus a surrogate for scalable inference is not a routine step in prior TMD work. The workflow makes sense for handling the computational cost at N3LO plus N4LL resummation. Replacing repeated theory evaluations with a trained emulator is a standard move when the underlying calculation is expensive, and it lets them do full posterior sampling over the free parameters. The global fit scope and the direct comparison to replicas are useful additions. The soft spot is validation. The final TMDs and their uncertainties stand or fall on whether the surrogate reproduces the true cross sections accurately across the sampled space and whether the AI-ranked forms avoid systematic bias in describing the data. The abstract does not include those metrics, so the strength of the extracted distributions will depend on how well those checks hold up in the full text. This is for readers working on TMD phenomenology and global parton distribution fits who need practical ways to scale uncertainty quantification as data and perturbative orders increase. It deserves a serious referee because the method is a clear extension of existing Bayesian TMD analyses and the application is relevant for collider precision work. I would send it out for review.

Referee Report

2 major / 3 minor

Summary. The manuscript presents a Bayesian extraction of unpolarized quark TMD PDFs from a global set of Drell-Yan data (fixed-target, RHIC, LHC) at N³LO perturbative accuracy combined with N⁴LL resummation. AI is used in two auxiliary roles: an iterative ranking of candidate functional forms for the nonperturbative TMD PDF at the initial scale and for the Collins-Soper kernel, and the training of a machine-learning surrogate emulator that replaces repeated cross-section evaluations inside an affine-invariant MCMC sampler. The resulting posteriors are compared with those obtained via the replica method, with emphasis on differences in the reported uncertainty estimates.

Significance. If the surrogate validation and functional-form selection are shown to be robust, the work offers a practical route to scalable, uncertainty-quantified TMD fits at high perturbative order. The explicit comparison between MCMC and replica uncertainties, together with the use of physics-informed constraints during AI ranking, provides a concrete test of how auxiliary machine-learning steps affect final TMD extractions; this is a useful methodological contribution to the field.

major comments (2)

[§4.3] §4.3 (surrogate validation): the manuscript must report quantitative accuracy metrics (e.g., maximum relative error, RMS error on a held-out test set spanning the full prior volume) and demonstrate that the emulator error is sub-dominant to the experimental uncertainties used in the likelihood; without these numbers the claim that the MCMC posteriors are reliable cannot be assessed.
[§3.1] §3.1 (AI ranking procedure): the iterative χ²-based ranking of nonperturbative ansätze should be supplemented by an explicit check that the selected forms remain stable when the data set is varied (e.g., leave-one-experiment-out tests) and that the physics constraints do not inadvertently exclude viable parameter regions; otherwise the model dependence introduced by the AI step remains unquantified.

minor comments (3)

[Table 2] Table 2: the reported χ²/dof values for the final MCMC fit should be accompanied by the number of data points and the effective number of parameters to allow direct comparison with replica results.
[Figure 7] Figure 7: the uncertainty bands on the extracted TMDs at different scales should be overlaid with at least one previous global extraction (e.g., from the literature cited in §1) for visual assessment of consistency.
The notation for the Collins-Soper kernel parameters should be unified between the AI-ranking section and the MCMC parameter list to avoid reader confusion.

Simulated Author's Rebuttal

2 responses · 0 unresolved

We thank the referee for the careful review and the recommendation for minor revision. The comments identify key areas where additional documentation will strengthen the manuscript, and we address each point below.

read point-by-point responses

Referee: [§4.3] §4.3 (surrogate validation): the manuscript must report quantitative accuracy metrics (e.g., maximum relative error, RMS error on a held-out test set spanning the full prior volume) and demonstrate that the emulator error is sub-dominant to the experimental uncertainties used in the likelihood; without these numbers the claim that the MCMC posteriors are reliable cannot be assessed.

Authors: We agree that quantitative validation metrics are required to substantiate the surrogate emulator. In the revised manuscript we will report the maximum relative error and RMS error on a held-out test set spanning the full prior volume. We have verified that the emulator error is sub-dominant to the experimental uncertainties entering the likelihood; the updated §4.3 will include these numbers and the corresponding comparison. revision: yes
Referee: [§3.1] §3.1 (AI ranking procedure): the iterative χ²-based ranking of nonperturbative ansätze should be supplemented by an explicit check that the selected forms remain stable when the data set is varied (e.g., leave-one-experiment-out tests) and that the physics constraints do not inadvertently exclude viable parameter regions; otherwise the model dependence introduced by the AI step remains unquantified.

Authors: We acknowledge that stability under data-set variations must be shown explicitly. The revised §3.1 will include leave-one-experiment-out tests confirming that the selected functional forms remain stable. We will also expand the discussion of the physics constraints to demonstrate that they enforce standard positivity and theoretical requirements without excluding viable parameter regions, thereby quantifying the model dependence introduced by the AI ranking. revision: yes

Circularity Check

0 steps flagged

No significant circularity detected in derivation chain

full rationale

The paper performs a standard Bayesian parameter extraction of TMD PDFs from external Drell-Yan datasets at N3LO+N4LL accuracy. AI components are used only for auxiliary tasks: iterative ranking of candidate nonperturbative functional forms via chi-squared minimization against data, and training of an ML surrogate emulator to accelerate repeated cross-section evaluations inside MCMC sampling. Neither step defines the target TMD parameters in terms of themselves, renames a fitted quantity as a prediction, or imports a uniqueness result via self-citation. The extracted posteriors remain directly constrained by the experimental measurements, with the AI tools serving solely to render the high-dimensional sampling computationally tractable; the central result is therefore independent of its own inputs.

Axiom & Free-Parameter Ledger

2 free parameters · 2 axioms · 0 invented entities

The extraction rests on standard perturbative QCD calculations at N3LO+N4LL, the assumption that Drell-Yan data constrain TMDs, and a set of free parameters in the nonperturbative model and evolution kernel that are determined by the Bayesian fit.

free parameters (2)

nonperturbative TMD PDF parameters at initial scale
Fitted to data via Bayesian inference; exact number and values not stated in abstract
Collins-Soper evolution kernel parameters
Fitted to data via Bayesian inference; exact number and values not stated in abstract

axioms (2)

domain assumption Perturbative QCD factorization for TMD cross sections holds at N3LO+N4LL accuracy
Invoked to justify the perturbative component of the calculation
domain assumption Drell-Yan measurements from fixed-target, RHIC, and LHC experiments can be combined in a global fit
Standard assumption for TMD global analyses

pith-pipeline@v0.9.0 · 5515 in / 1413 out tokens · 31339 ms · 2026-05-10T12:55:35.764540+00:00 · methodology

discussion (0)

Forward citations

Cited by 3 Pith papers

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score.

TMDs in the Lens of Generative AI: A Pixel-Based Approach to Partonic Imaging
hep-ph 2026-05 unverdicted novelty 7.0

A nonparametric pixel-based Bayesian method integrates TMD evolution with generative AI and SVD to image parton distributions and reveal null TMDs unconstrained by observables.
Simplified approach to extracting nucleon transversity in collinear factorization using near-side energy-energy correlators
hep-ph 2026-04 unverdicted novelty 7.0

A new method extracts the nucleon transversity PDF via near-side energy-energy correlators in dihadron fragmentation under collinear factorization, with leading-order results for SIDIS and e+e- annihilation that resem...
Simplified approach to extracting nucleon transversity in collinear factorization using near-side energy-energy correlators
hep-ph 2026-04 unverdicted novelty 7.0

A new approach using near-side energy-energy correlators in dihadron fragmentation enables extraction of nucleon transversity PDF in collinear factorization without modeling intrinsic transverse momentum or dihadron r...

Reference graph

Works this paper leans on

75 extracted references · 58 canonical work pages · cited by 2 Pith papers · 5 internal anchors

[1]

Electron Ion Collider: The Next QCD Frontier - Understanding the glue that binds us all

A. Accardiet. al.,Electron Ion Collider: The Next QCD Frontier: Understanding the glue that binds us all,Eur. Phys. J. A52(2016) no. 9 268 [arXiv:1212.1701 [nucl-ex]]

work page Pith review arXiv 2016
[2]

Collins,Foundations of Perturbative QCD, vol

J. Collins,Foundations of Perturbative QCD, vol. 32. Cambridge University Press, 2011

2011
[3]

TMD Handbook,

R. Boussarieet. al.,TMD Handbook,arXiv:2304.03302 [hep-ph]

work page arXiv
[4]

Science Requirements and Detector Concepts for the Electron-Ion Collider: EIC Yellow Report

R. Abdul Khaleket. al.,Science Requirements and Detector Concepts for the Electron-Ion Collider: EIC Yellow Report,Nucl. Phys. A1026(2022) 122447 [arXiv:2103.05419 [physics.ins-det]]

work page internal anchor Pith review arXiv 2022
[5]

Precision QCD with the Electron-Ion Collider

C. Alexandrouet. al.,Precision QCD with the Electron-Ion Collider,arXiv:2604.04765 [hep-ph]

work page internal anchor Pith review Pith/arXiv arXiv
[6]

V. Moos, I. Scimemi, A. Vladimirov and P. Zurita,Determination of unpolarized TMD distributions from the fit of Drell-Yan and SIDIS data at N 4LL,JHEP11(2025) 134 [arXiv:2503.11201 [hep-ph]]. [7]MAP (Multi-dimensional Analyses of Partonic distributions)collaboration, A. Bacchetta, V. Bertone, C. Bissolotti, G. Bozzi, M. Cerutti, F. Delcarro, M. Radici, L....

work page arXiv 2025
[7]

P. C. Barryet. al.,First simultaneous analysis of transverse momentum dependent and collinear parton distributions in the proton,arXiv:2510.13771 [hep-ph]. [9]MAP (Multi-dimensional Analyses of Partonic distributions)collaboration, A. Bacchetta, V. Bertone, C. Bissolotti, M. Cerutti, M. Radici, S. Rodini and L. Rossi, Neural-Network Extraction of Unpolari...

work page arXiv 2025
[8]

Avkhadiev, V

A. Avkhadiev, V. Bertone, C. Bissolotti, M. Cerutti, Y. Fu, S. Rodini, P. Shanahan, M. Wagman and Y. Zhao,An extraction of the Collins-Soper kernel from a joint analysis of experimental and lattice data,arXiv:2510.26489 [hep-ph]. [11]MAP (Multi-dimensional Analyses of Partonic distributions)collaboration, A. Bacchetta, A. Bongallino, M. Cerutti, M. Radici...

work page arXiv 2025
[9]

M. G. Echevarria, Z.-B. Kang and J. Terry,Global analysis of the Sivers functions at NLO+NNLL in QCD,JHEP01(2021) 126 [arXiv:2009.10710 [hep-ph]]

work page arXiv 2021
[10]

Alrashed, D

M. Alrashed, D. Anderle, Z.-B. Kang, J. Terry and H. Xing,Three-dimensional imaging in nuclei,Phys. Rev. Lett.129(2022) no. 24 242001 [arXiv:2107.12401 [hep-ph]]

work page arXiv 2022
[11]

Candido, L

A. Candido, L. Del Debbio, T. Giani and G. Petrillo,Bayesian inference with Gaussian processes for the determination of parton distribution functions,Eur. Phys. J. C84 (2024) no. 7 716 [arXiv:2404.07573 [hep-ph]]

work page arXiv 2024
[12]

Capel, R

F. Capel, R. Aggarwal, M. Botje, A. Caldwell, O. Schulz and A. Verbytskyi,Novel parton density determination code,Phys. Rev. D110(2024) no. 1 014024 [arXiv:2401.17729 [hep-ph]]

work page arXiv 2024
[13]

Casuga, H

C. Casuga, H. H¨ anninen and H. M¨ antysaari,Initial condition for the Balitsky-Kovchegov – 41 – equation at next-to-leading order,Phys. Rev. D112(2025) no. 3 034003 [arXiv:2506.00487 [hep-ph]]

work page arXiv 2025
[14]

M¨ antysaari, H

H. M¨ antysaari, H. Roch, F. Salazar, B. Schenke, C. Shen and W. Zhao,Global Bayesian analysis of J/ψphotoproduction on proton and lead targets,Phys. Rev. D113(2026) no. 1 014038 [arXiv:2507.14087 [hep-ph]]

work page arXiv 2026
[15]

J. E. Bernhard, J. S. Moreland, S. A. Bass, J. Liu and U. Heinz,Applying bayesian parameter estimation to relativistic heavy-ion collisions: Simultaneous characterization of the initial state and quark-gluon plasma medium,Phys. Rev. C94(Aug, 2016) 024907. https://link.aps.org/doi/10.1103/PhysRevC.94.024907. [19]JETSCAPEcollaboration, A. Mankolliet. al.,Lo...

work page doi:10.1103/physrevc.94.024907 2016
[16]

Landry and R

P. Landry and R. Essick,Nonparametric inference of the neutron star equation of state from gravitational wave observations,Phys. Rev. D99(Apr, 2019) 084049. https://link.aps.org/doi/10.1103/PhysRevD.99.084049

work page doi:10.1103/physrevd.99.084049 2019
[17]

Brandes and W

L. Brandes and W. Weise,Implications of latest nicer data for the neutron star equation of state,Phys. Rev. D111(Feb, 2025) 034005. https://link.aps.org/doi/10.1103/PhysRevD.111.034005

work page doi:10.1103/physrevd.111.034005 2025
[18]

S. Huth, P. T. H. Pang, I. Tewset. al.,Constraining neutron-star matter with microscopic and macroscopic collisions,Nature606(2022) 276

2022
[19]

Efron and R

B. Efron and R. J. Tibshirani,An Introduction to the Bootstrap. Chapman and Hall/CRC, 1 ed., 1994

1994
[20]

A. M. Stuart,Inverse problems: A bayesian perspective,Acta Numerica19(2010) 451–559

2010
[21]

R. E. Kass, L. Tierney and J. J. B. Kadane 1990

1990
[22]

Scimemi and A

I. Scimemi and A. Vladimirov,Analysis of vector boson production within TMD factorization,Eur. Phys. J. C78(2018) no. 2 89 [arXiv:1706.01473 [hep-ph]]

work page arXiv 2018
[23]

Scimemi and A

I. Scimemi and A. Vladimirov,Non-perturbative structure of semi-inclusive deep-inelastic and Drell-Yan scattering at small transverse momentum,JHEP06(2020) 137 [arXiv:1912.06532 [hep-ph]]

work page arXiv 2020
[24]

Piloneta and A

S. Piloneta and A. Vladimirov,Angular distributions of Drell-Yan leptons in the TMD factorization approach,JHEP12(2024) 059 [arXiv:2407.06277 [hep-ph]]

work page arXiv 2024
[25]

Neumann and J

T. Neumann and J. Campbell,Fiducial Drell-Yan production at the LHC improved by transverse-momentum resummation at N4LLp+N3LO,Phys. Rev. D107(2023) no. 1 L011506 [arXiv:2207.07056 [hep-ph]]

work page arXiv 2023
[26]

A. A. Vladimirov,Correspondence between Soft and Rapidity Anomalous Dimensions,Phys. Rev. Lett.118(2017) no. 6 062001 [arXiv:1610.05791 [hep-ph]]

work page arXiv 2017
[27]

Vladimirov,Structure of rapidity divergences in multi-parton scattering soft factors, JHEP04(2018) 045, [1707.07606]

A. Vladimirov,Structure of rapidity divergences in multi-parton scattering soft factors,JHEP 04(2018) 045 [arXiv:1707.07606 [hep-ph]]

work page arXiv 2018
[28]

LHAPDF6: parton density access in the LHC precision era

A. Buckley, J. Ferrando, S. Lloyd, K. Nordstr¨ om, B. Page, M. R¨ ufenacht, M. Sch¨ onherr and G. Watt,LHAPDF6: parton density access in the LHC precision era,Eur. Phys. J. C75 (2015) 132 [arXiv:1412.7420 [hep-ph]]. – 42 –

work page Pith review arXiv 2015
[29]

J. C. Collins, D. E. Soper and G. F. Sterman,Transverse Momentum Distribution in Drell-Yan Pair and W and Z Boson Production,Nucl. Phys. B250(1985) 199

1985
[30]

Becher and M

T. Becher and M. Neubert,Drell-Yan Production at Smallq T , Transverse Parton Distributions and the Collinear Anomaly,Eur. Phys. J. C71(2011) 1665 [arXiv:1007.4005 [hep-ph]]

work page arXiv 2011
[31]

M. G. Echevarria, A. Idilbi and I. Scimemi,Factorization Theorem For Drell-Yan At Lowq T And Transverse Momentum Distributions On-The-Light-Cone,JHEP07(2012) 002 [arXiv:1111.4996 [hep-ph]]

work page arXiv 2012
[32]

M. G. Echevarria, I. Scimemi and A. Vladimirov,Unpolarized Transverse Momentum Dependent Parton Distribution and Fragmentation Functions at next-to-next-to-leading order,JHEP09(2016) 004 [arXiv:1604.07869 [hep-ph]]

work page arXiv 2016
[33]

Gehrmann, T

T. Gehrmann, T. Lubbert and L. L. Yang,Transverse parton distribution functions at next-to-next-to-leading order: the quark-to-quark case,Phys. Rev. Lett.109(2012) 242003 [arXiv:1209.0682 [hep-ph]]

work page arXiv 2012
[34]

Gehrmann, T

T. Gehrmann, T. Luebbert and L. L. Yang,Calculation of the transverse parton distribution functions at next-to-next-to-leading order,JHEP06(2014) 155 [arXiv:1403.6451 [hep-ph]]

work page arXiv 2014
[35]

J. C. Collins and D. E. Soper,Back-To-Back Jets in QCD,Nucl. Phys. B193(1981) 381. [Erratum: Nucl.Phys.B 213, 545 (1983)]

1981
[36]

A. A. Vladimirov,Self-contained definition of the Collins-Soper kernel,Phys. Rev. Lett.125 (2020) no. 19 192002 [arXiv:2003.02288 [hep-ph]]

work page arXiv 2020
[37]

G. P. Korchemsky and A. V. Radyushkin,Loop Space Formalism and Renormalization Group for the Infrared Asymptotics of QCD,Phys. Lett. B171(1986) 459

1986
[38]

Becher, A

T. Becher, A. Broggio and A. Ferroglia,Introduction to Soft-Collinear Effective Theory, vol. 896. Springer, 2015

2015
[39]

Li and H

Y. Li and H. X. Zhu,Bootstrapping Rapidity Anomalous Dimensions for Transverse-Momentum Resummation,Phys. Rev. Lett.118(2017) no. 2 022004 [arXiv:1604.01404 [hep-ph]]

work page arXiv 2017
[40]

Moult, H

I. Moult, H. X. Zhu and Y. J. Zhu,The four loop QCD rapidity anomalous dimension,JHEP 08(2022) 280 [arXiv:2205.02249 [hep-ph]]

work page arXiv 2022
[41]

C. Duhr, B. Mistlberger and G. Vita,Four-Loop Rapidity Anomalous Dimension and Event Shapes to Fourth Logarithmic Order,Phys. Rev. Lett.129(2022) no. 16 162001 [arXiv:2205.02242 [hep-ph]]

work page arXiv 2022
[42]

S. Moch, J. A. M. Vermaseren and A. Vogt,Three-loop results for quark and gluon form-factors,Phys. Lett. B625(2005) 245 [arXiv:hep-ph/0508055]

work page arXiv 2005
[43]

Gehrmann, E

T. Gehrmann, E. W. N. Glover, T. Huber, N. Ikizlerli and C. Studerus,Calculation of the quark and gluon form factors to three loops in QCD,JHEP06(2010) 094 [arXiv:1004.3653 [hep-ph]]

work page arXiv 2010
[44]

M. A. Ebert, B. Mistlberger and G. Vita,Transverse momentum dependent PDFs at N 3LO, JHEP09(2020) 146 [arXiv:2006.05329 [hep-ph]]

work page arXiv 2020
[45]

Luo, T.-Z

M.-x. Luo, T.-Z. Yang, H. X. Zhu and Y. J. Zhu,Unpolarized quark and gluon TMD PDFs and FFs at N 3LO,JHEP06(2021) 115 [arXiv:2012.03256 [hep-ph]]. – 43 –

work page arXiv 2021
[46]

Agarwal, A

B. Agarwal, A. von Manteuffel, E. Panzer and R. M. Schabinger,Four-loop collinear anomalous dimensions in QCD and N=4 super Yang-Mills,Phys. Lett. B820(2021) 136503 [arXiv:2102.09725 [hep-ph]]

work page arXiv 2021
[47]

M. G. Echevarria, I. Scimemi and A. Vladimirov,Universal transverse momentum dependent soft function at NNLO,Phys. Rev. D93(2016) no. 5 054004 [arXiv:1511.05590 [hep-ph]]

work page arXiv 2016
[48]

J. M. Henn, G. P. Korchemsky and B. Mistlberger,The full four-loop cusp anomalous dimension inN= 4super Yang-Mills and QCD,JHEP04(2020) 018 [arXiv:1911.10174 [hep-th]]

work page arXiv 2020
[49]

Herzog, S

F. Herzog, S. Moch, B. Ruijl, T. Ueda, J. A. M. Vermaseren and A. Vogt,Five-loop contributions to low-N non-singlet anomalous dimensions in QCD,Phys. Lett. B790(2019) 436 [arXiv:1812.11818 [hep-ph]]

work page arXiv 2019
[50]

V. Moos, I. Scimemi, A. Vladimirov and P. Zurita,Extraction of unpolarized transverse momentum distributions from the fit of Drell-Yan data at N 4LL,JHEP05(2024) 036 [arXiv:2305.07473 [hep-ph]]

work page arXiv 2024
[51]

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. C81(2021) no. 4 341 [arXiv:2012.04684 [hep-ph]]

work page arXiv 2021
[52]

McGowan, T

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

work page arXiv 2023
[53]

Fazio,Preliminary STARZ-boson transverse-momentum spectrum at √s= 510GeV, 7,

S. Fazio,Preliminary STARZ-boson transverse-momentum spectrum at √s= 510GeV, 7,
[54]

[58]CDFcollaboration, T

Collaboration-shared preliminary result provided by email on behalf of the STAR Collaboration, July 18, 2019. [58]CDFcollaboration, T. Affolderet. al.,The transverse momentum and total cross section of e+e− pairs in theZboson region fromp¯pcollisions at √s= 1.8TeV,Phys. Rev. Lett.84 (2000) 845 [arXiv:hep-ex/0001021]. [59]CDFcollaboration, T. Aaltonenet. a...

work page arXiv 2019
[55]

A. S. Itoet. al.,Measurement of the Continuum of Dimuons Produced in High-Energy Proton - Nucleus Collisions,Phys. Rev. D23(1981) 604

1981
[56]

Morenoet

G. Morenoet. al.,Dimuon Production in Proton - Copper Collisions at √s= 38.8-GeV, Phys. Rev. D43(1991) 2815. [73]E772collaboration, P. L. McGaugheyet. al.,Cross-sections for the production of high mass muon pairs from 800-GeV proton bombardment of H-2,Phys. Rev. D50(1994) 3038. [Erratum: Phys.Rev.D 60, 119903 (1999)]

1991
[57]

T.-J. Houet. al.,Reconstruction of Monte Carlo replicas from Hessian parton distributions, JHEP03(2017) 099 [arXiv:1607.06066 [hep-ph]]. [75]MAP (Multi-dimensional Analyses of Partonic distributions)collaboration, A. Bacchetta, V. Bertone, C. Bissolotti, G. Bozzi, M. Cerutti, F. Piacenza, M. Radici and A. Signori,Unpolarized transverse momentum distributi...

work page arXiv 2017
[58]

M. Chen, J. Tworek, H. Jun, Q. Yuan, H. Ponde de Oliveira Pinto, J. Kaplan, H. Edwards, Y. Burda, N. Joseph, G. Brockmanet. al.,Evaluating Large Language Models Trained on Code,arXiv:2107.03374 [cs.LG]

work page internal anchor Pith review Pith/arXiv arXiv
[59]

Introducing GPT-5.4

OpenAI, “Introducing GPT-5.4.” urlhttps://openai.com/index/introducing-gpt-5-4/, 3, 2026. GPT-5.4 released in ChatGPT, the API, and Codex. [78]NNPDFcollaboration, R. D. Ball, L. Del Debbio, S. Forte, A. Guffanti, J. I. Latorre, J. Rojo and M. Ubiali,Fitting Parton Distribution Data with Multiplicative Normalization Uncertainties,JHEP05(2010) 075 [arXiv:09...

work page arXiv 2026
[60]

emcee: The MCMC Hammer

D. Foreman-Mackey, D. W. Hogg, D. Lang and J. Goodman,emcee: The MCMC Hammer, Publ. Astron. Soc. Pac.125(2013) 306 [arXiv:1202.3665 [astro-ph.IM]]

work page internal anchor Pith review arXiv 2013
[61]

Goodman and J

J. Goodman and J. Weare,Ensemble samplers with affine invariance,Communications in Applied Mathematics and Computational Science5(2010) no. 1 65

2010
[62]

M. D. McKay, R. J. Beckman and W. J. Conover,A comparison of three methods for selecting values of input variables in the analysis of output from a computer code, Technometrics21(1979) no. 2 239

1979
[63]

I. T. Jolliffe and J. Cadima,Principal component analysis: a review and recent developments, Philos. Trans. Roy. Soc. A374(2016) no. 2065 20150202

2016
[64]

Simple and Scalable Predictive Uncertainty Estimation using Deep Ensembles

B. Lakshminarayanan, A. Pritzel and C. Blundell inAdvances in Neural Information Processing Systems 30, pp. 6402–6413, 2017.arXiv:1612.01474 [cs.LG]

work page Pith review arXiv 2017
[65]

Decoupled Weight Decay Regularization

I. Loshchilov and F. Hutter in7th International Conference on Learning Representations (ICLR 2019), 2019.arXiv:1711.05101 [cs.LG]

work page internal anchor Pith review Pith/arXiv arXiv 2019
[66]

SGDR: Stochastic Gradient Descent with Warm Restarts

I. Loshchilov and F. Hutter in5th International Conference on Learning Representations (ICLR 2017), 2017.arXiv:1608.03983 [cs.LG]

work page Pith review arXiv 2017
[67]

C. J. F. ter Braak,A Markov Chain Monte Carlo version of the genetic algorithm Differential Evolution: easy Bayesian computing for real parameter spaces,Statistics and Computing16(2006) no. 3 239

2006
[68]

C. J. F. ter Braak and J. A. Vrugt,Differential Evolution Markov Chain with snooker updater and fewer chains,Statistics and Computing18(2008) no. 4 435

2008
[69]

Vehtari, A

A. Vehtari, A. Gelman, D. Simpson, B. Carpenter and P.-C. B¨ urkner,Rank-normalization, folding, and localization: An improved ˆRfor assessing convergence of mcmc,Bayesian Analysis16(2021) no. 2 667 [arXiv:1903.08008 [stat.CO]]

work page arXiv 2021
[70]

Z.-B. Kang, J. Penttala and C. Zhang,Determination of the strong coupling constant and the Collins-Soper kernel from the energy-energy correlator ine +e− collisions,arXiv:2410.21435 [hep-ph]

work page arXiv
[71]

Bollweg, X

D. Bollweg, X. Gao, S. Mukherjee and Y. Zhao,Nonperturbative Collins-Soper kernel from chiral quarks with physical masses,Phys. Lett. B852(2024) 138617 [arXiv:2403.00664 [hep-lat]]. [91]Lattice Parton (LPC)collaboration, M.-H. Chuet. al.,Lattice calculation of the intrinsic soft function and the Collins-Soper kernel,JHEP08(2023) 172 [arXiv:2306.06488 [hep-lat]]

work page arXiv 2024
[72]

Avkhadiev, P

A. Avkhadiev, P. E. Shanahan, M. L. Wagman and Y. Zhao,Determination of the Collins-Soper Kernel from Lattice QCD,Phys. Rev. Lett.132(2024) no. 23 231901 [arXiv:2402.06725 [hep-lat]]

work page arXiv 2024
[73]

Del Debbio, T

L. Del Debbio, T. Giani and M. Wilson,Bayesian approach to inverse problems: an application to nnpdf closure testing,European Physical Journal C82(2022) no. 4 330. https://doi.org/10.1140/epjc/s10052-022-10297-x

work page doi:10.1140/epjc/s10052-022-10297-x 2022
[74]

Watt and R

G. Watt and R. S. Thorne,Study of Monte Carlo approach to experimental uncertainty propagation with MSTW 2008 PDFs,JHEP08(2012) 052 [arXiv:1205.4024 [hep-ph]]

work page arXiv 2008
[75]

M. N. Costantini, M. Madigan, L. Mantani and J. M. Moore,A critical study of the monte – 46 – carlo replica method,Journal of High Energy Physics2024(2024) no. 12 64. https://doi.org/10.1007/JHEP12(2024)064. – 47 –

work page doi:10.1007/jhep12(2024)064 2024