Analytic parametrisations of the scale dependence of proton parton distributions
Listen
Radio episode about this paper
Transcript
Introduction to the show: ident: High Energy Physics: Phenomenology. The day in particle phenomenology: models, signatures and constraints.
Kai: I'm Kai.
Mira: And I'm Mira, here with Lev.
Kai: We're looking at the paper: "Analytic parametrisations of the scale dependence of proton parton distributions".
Mira: Here's the gist of the paper. Proton parton distribution functions are non-perturbative inputs whose scale dependence is governed by the Dokshitzer–Gribov–Lipatov–Altarelli–Parisi (DGLAP) equations, and this paper presents an analytic representation of collinear proton PDFs obtained by promoting the parameters of a standard x-space parametrisation to functions of the scale The resulting ML-PDFs reproduce integrated distributions at the subpercent level for most flavours, with deviations reaching about 1.5% for the strange quark at the highest scales Analytic Representation and Fitting Procedure Proton PDFs are commonly parametrised at an initial scale Q 2 0 using a functional form that captures their small- and large-x behaviour, supplemented by a factor that provides additional flexibility A representative form is f i(x, Q 2 0) = N i x alpha i (1 - x) beta i P(x, c ij), where i denotes the parton flavour, x is the fraction of the proton’s longitudinal momentum carried by the parton, and N i is a normalisation factor The aim is to construct an explicit analytic representation in (x, Q 2) that can be evaluated without interpolating PDF grids The coefficients are determined in two steps
4: First, the HERAPDF2.0 NLO set
2: is sampled at O (5000) randomly selected points with x ∈
10−4, 0.65: and Q ∈
Qmin, 1000: GeV The x dependence is fitted at fixed Q to determine the coefficients Second, the Mathematica routine FindFormula suggests functional forms for their scale dependence, which are used to construct the final analytic parametrisations
4: Distribution Accuracy and Consistency Checks The resulting analytic approximations, denoted ML-PDFs, reproduce the output of DGLAP evolution embodied in the reference set; they are not new analytic solutions of the evolution equations The integral error Δ i(Q 2) = I
f ML i, Q 2: − I
f HERA i, Q 2: / I
f HERA i, Q 2: measures differences between integrated distributions and is less sensitive than point-wise relative errors to fluctuations where the PDF is small Most distributions have deviations below 0.5%, while charm reaches roughly 0.7% and the strange-quark distribution exhibits the largest deviation, approximately 1.5% at Q ≈ 1000 GeV The valence and momentum sum rules hold at the percent level, and the two hadronic benchmarks agree with the reference predictions within about 2% and 5% in the respective regions considered Hadronic Observables and Computational Performance The predictions obtained with ML-PDFs are compared with those using the original HERAPDF2.0 grids in Fig.
Kai: 3 For single-inclusive pion production, the differences remain well below 2% and are mostly contained within the HERAPDF2.0 uncertainty band For photon–pion production, the deviations are within 5% for pT ≤ 10 GeV, where the cross section is largest The execution times obtained for both processes are summarised in Table 1 Relative to LHAPDF, the analytic ML-PDFs reduce the total run time by approximately 11% for p p → π and 30% for p p → γπ
4: We have presented an explicit analytic approximation to the (x, Q 2) dependence of collinear proton PDFs Most integrated distributions agree with the reference set at the subpercent level, while the strange-quark deviation reaches approximately 1.5% at Q ≈ 1000 GeV The valence and momentum sum rules hold at the percent level, and the two hadronic benchmarks agree with the reference predictions within about 2% and 5% in the respective regions considered Direct ML-PDF evaluation also reduces the total Monte Carlo run time by 11% and 30% in these benchmarks The same strategy could be explored for other PDF sets and fragmentation functions REFERENCES
1: G.
Mira: Altarelli and G.
Kai: Parisi, Asymptotic freedom in parton language, Nucl.
Paper summary: Kai: Right. And what they claim is that by taking the parameters from a standard x space parametrisation and making them functions of the scale itself, they can get an analytic expression that you don't have to interpolate for >
Mira: That's what David F. Rentería-Estrada and his team did. They used machine-assisted fitting techniques on the HERAPDF2 point 0 NLO set, sampling points in x from ten to negative four up to zero point six five and scales all the way up to one tera electron volt > <ref:2611.08272#pg1,the HERAPDF2.0 NLO set>
Lev: So they are using a standard set as a reference point for their fitting, which makes sense for getting those coefficients right before you even look at the scale dependence part >
Kai: Exactly. The resulting ML-PDFs they create reproduce integrated distributions at the subpercent level for most of the flavors, but there's a deviation reaching about one point five percent specifically for the strange quark as they go to the highest scales, around one thousand electron volt >
Mira: That one point five percent deviation is worth watching because it points to where their analytic form might need more refinement at those higher energies >
Lev: From an engineering side, getting that level of accuracy across such a wide range of x and Q two without the heavy computational cost of grid interpolation sounds like a real win for simulation times >
Conclusion: Kai: So we're wrapping up this look at "Analytic parametrisations of the scale dependence of proton parton distributions". The core idea is moving from grids to an explicit analytic description that accounts for the scale change >
Mira: It seems like the authors are focused on how much more accurate this direct representation is compared to simply interpolating those existing HERAPDF2 point 0 grids > <ref:2611.08272#pg1>
Lev: What this means practically, if you're running Monte Carlo simulations for something like high energy colliders, having an analytic expression instead of an interpolation routine could seriously cut down the total run time >
Kai: And the paper shows that for single inclusive pion production and photon to pion production, those differences stay well below two percent >
Mira: But they also flagged that this direct analytic method isn't a new solution to the DGLAP evolution equations; it's an approximation based on fitting existing data >
Lev: And they did admit there are cutoffs for the gluon and other heavy quarks, like a step function, which is important because those choices influence how the result behaves in regions where x is small >
Instituto de Fsa Corpuscular, Universitat de Valncia--Consejo Superior de Investigaciones Cientfica
hep-ph
Submitted: 2026-11-05
Updated: 2026-11-05
Comments: 6 pages, 3 figures, 1 table, contribution to the proceedings of the 43rd International Conference on High Energy Physics (ICHEP 2026), 30 July - 5 August 2026, Natal, Brazil
License: http://creativecommons.org/licenses/by/4.0/
Importance score: 77/100
The gist: Proton parton distribution functions are non-perturbative inputs whose scale dependence is governed by the Dokshitzer–Gribov–Lipatov–Altarelli–Parisi (DGLAP) equations…
Key concepts
- Proton Parton Distribution Functions (PDFs)
- These are non-perturbative inputs describing the probability of finding a specific type of parton (like a quark or gluon) carrying a certain fraction of the proton's total momentum. Their behavior changes depending on the energy scale at which they are measured.
- DGLAP Equations
- These equations govern how PDFs evolve with respect to the energy scale, known as DGLAP evolution. They describe how the parton distributions change as you probe them at different levels of resolution.
- Analytic Representation
- This involves creating a direct mathematical formula for PDFs that depends on both the momentum fraction (x) and the energy scale ($Q^2$). Instead of using pre-calculated grids, this approach allows PDFs to be evaluated directly using the analytic formula.
Terminology
Summary
Proton parton distribution functions are non-perturbative inputs whose scale dependence is governed by the Dokshitzer–Gribov–Lipatov–Altarelli–Parisi (DGLAP) equations, and this paper presents an analytic representation of collinear proton PDFs obtained by promoting the parameters of a standard x-space parametrisation to functions of the scale The resulting ML-PDFs reproduce integrated distributions at the subpercent level for most flavours, with deviations reaching about 1. p. 0 renteria@ific. es
Parton distribution functions (PDFs) are non-perturbative inputs whose scale dependence is governed by the Dokshitzer–Gribov–Lipatov–Altarelli–Parisi (DGLAP) equations.Where the paper says it
Analytic Representation and Fitting Procedure
Proton PDFs are commonly parametrised at an initial scale Q 2 0 using a functional form that captures their small- and large-x behaviour, supplemented by a factor that provides additional flexibility A representative form is f i(x, Q 2 0) = N i x alpha i (1 - x) beta i P(x, c ij), where i denotes the parton flavour, x is the fraction of the proton’s longitudinal momentum carried by the parton, and N i is a normalisation factor The aim is to construct an explicit analytic representation in (x, Q 2) that can be evaluated without interpolating PDF grids The coefficients are determined in two steps [4] First, the HERAPDF2.0 NLO set [2] is sampled at O (5000) randomly selected points with x ∈ [10−4, 0. p. 1 First, the HERAPDF2. 0 NLO set [2] is sampled at O (5000) randomly selected points
with 𝑥 ∈ [10−4
, 0.Where the paper says it
Distribution Accuracy and Consistency Checks
The resulting analytic approximations, denoted ML-PDFs, reproduce the output of DGLAP evolution embodied in the reference set; they are not new analytic solutions of the evolution equations The integral error Δ i(Q 2) = I[f ML i, Q 2] − I[f HERA i, Q 2] / I[f HERA i, Q 2] measures differences between integrated distributions and is less sensitive than point-wise relative errors to fluctuations where the PDF is small Most distributions have deviations below 0.5%, while charm reaches roughly 0. p. 3 Most distributions have deviations below 0. 5%, while charm reaches roughly 0. 5%, while charm reaches roughly 0. 7% and the
strange-quark distribution exhibits the largest deviation, approximately 1.Where the paper says it
Where the paper says it
Hadronic Observables and Computational Performance
The predictions obtained with ML-PDFs are compared with those using the original HERAPDF2.0 grids in Fig. p. 3 The predictions obtained with ML-PDFs are
compared with those using the original HERAPDF2. 0 grids in Fig. 0 grids in Fig. For single-inclusive pion production, the differences remain well below 2% and are mostly contained within the HERAPDF2.Where the paper says it
Where the paper says it
Conclusion
We have presented an explicit analytic approximation to the (x, Q 2) dependence of collinear proton PDFs Most integrated distributions agree with the reference set at the subpercent level, while the strange-quark deviation reaches approximately 1.5% at Q ≈ 1000 GeV The valence and momentum sum rules hold at the percent level, and the two hadronic benchmarks agree with the reference predictions within about 2% and 5% in the respective regions considered Direct ML-PDF evaluation also reduces the total Monte Carlo run time by 11% and 30% in these benchmarks The same strategy could be explored for other PDF sets and fragmentation functions
REFERENCES
[1] G. Altarelli and G. p. 0 renteria@ific. es
Parton distribution functions (PDFs) are non-perturbative inputs whose scale dependence is governed by the Dokshitzer–Gribov–Lipatov–Altarelli–Parisi (DGLAP) equations. Altarelli and G. Parisi, Asymptotic freedom in parton language, Nucl.Where the paper says it
Where the paper says it
[2] H1 and ZEUS Collaborations, Combination of measurements of inclusive deep inelastic e+ p scattering cross sections and QCD analysis of HERA data, Eur. Phys. J. C 75 (2015) 580, arXiv:1506. p. 5 [2] H1 and ZEUS Collaborations, Combination of measurements of inclusive deep inelastic 𝑒
± 𝑝 scattering
cross sections and QCD analysis of HERA data, Eur. C 75 (2015) 580, arXiv:1506.Where the paper says it
[3] A. Buckley et al. p. 5 C 75 (2015) 580, arXiv:1506. Buckley et al. Buckley et al. , LHAPDF6: parton density access in the LHC precision era, Eur. , LHAPDF6: parton density access in the LHC precision era, Eur. C 75 (2015)
132, arXiv:1412.Where the paper says it
Where the paper says it
Where the paper says it
[4] S. A. Ochoa-Oregón, D. p. 5 C 75 (2015)
132, arXiv:1412. Ochoa-Oregón, D. C 75 (2015)
132, arXiv:1412. Ochoa-Oregón, D. Analytic scale dependence of PDFs David F. Rentería-Estrada
1. Analytic scale dependence of PDFs David F. Rentería-Estrada
1. Analytic scale dependence of PDFs David F. Rentería-Estrada
1. Analytic scale dependence of PDFs David F. Rentería-Estrada
1. Rentería-Estrada, R. Hernández-Pinto, G. Rentería-Estrada, R. Hernández-Pinto, G. Hernández-Pinto, G. Sborlini and
P.Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
[5] S. A. Ochoa-Oregón, D. F. Rentería-Estrada, R. J. Hernández-Pinto, G. F. R. Sborlini and P Zurita, Ancillary files for Zurita, Ancillary files for "Using analytic models to describe effective PDFs", Zenodo (2024),
doi:10. 5281/zenodo.Using analytic models to describe effective PDFs
, Zenodo (2024) doi:10.5281/zenodo. p. 5Where the paper says it
[6] D. de Florian, R. p. 5 5281/zenodo. de Florian, R. Stratmann, Parton-to-pion fragmentation reloaded, Phys. D 91 (2015) 014035, doi:10. D 110 (2024) 036019,
doi:10. 1103/PhysRevD.Where the paper says it
Where the paper says it
Where the paper says it
[7] D. de Florian and G F R Sborlini, Hadron plus photon production in polarized hadronic collisions at next-to-leading order accuracy, Phys Rev D 83 (2011) 074022 arXiv:1011. p. 5 Sborlini, Hadron plus photon production in polarized hadronic collisions at
next-to-leading order accuracy, Phys. D 83 (2011) 074022, arXiv:1011.Where the paper says it
[8] D. F. Rentería-Estrada, R. J. Hernández-Pinto and G F R Sborlini, Analysis of the internal structure of hadrons using direct photon production, Symmetry 13 (2021) 942 arXiv:2104. p. 5 Hernández-Pinto and G. Sborlini, Analysis of the internal structure
of hadrons using direct photon production, Symmetry 13 (2021) 942, arXiv:2104.Where the paper says it
[9] D. F. Rentería-Estrada, R. J. Hernández-Pinto, G Rodrigo, R Sassot and G F R Sborlini, A quantum representation of pion fragmentation functions through variational quantum circuits, arXiv:2609. p. 5 Sassot and G. Sborlini, A quantum
representation of 𝜋 fragmentation functions through variational quantum circuits, arXiv:2609.Where the paper says it
Green — the paper says it, and the page is printed. Red — no passage carries it, so treat that sentence as unverified. The same check runs on your own text: check a summary against its source, $5.
Improvements for AI systems
-
Improved AI can perform direct evaluation of parton distribution functions without interpolating tabulated grids, which reduces total Monte Carlo run time by about 11% for p p -> pi and 30% for p p -> gamma pi.
-
The system can construct explicit analytic representations of the x space dependence and scale dependence of collinear proton PDFs using a functional form like f i(x, Q2) = N i(Q2) x alpha i (Q2) (1 - x) beta i (Q2) P(x, c i j)(4).
-
The improved system can assess accuracy by comparing integral errors Delta i(Q2), defined as I[f ML i, Q2] - I[f HER A i, Q2], which is less sensitive than pointwise relative errors to fluctuations where the PDF is small.
-
The AI can validate phenomenological impact by comparing predictions of hadronic observables like single-inclusive pion production and photon–pion production against reference sets, showing deviations remain well below 2% and 5% respectively.
Abstract
Parton distribution functions (PDFs) are non-perturbative inputs whose scale dependence is governed by the Dokshitzer-Gribov-Lipatov-Altarelli-Parisi (DGLAP) equations. Their numerical evaluation commonly relies on interpolation of tabulated grids. We discuss an analytic representation of collinear proton PDFs in (x,Q 2), obtained by promoting the parameters of a standard x-space parametrisation to functions of the scale. The coefficients are reconstructed from the HERAPDF2.0 NLO set using machine-assisted fitting techniques, with sampling in 10-4 at most x at most 0.65 and scales up to 1 TeV. The resulting ML-PDFs reproduce integrated distributions at the sub-percent level for most flavours, with deviations reaching about 1.5% for the strange quark at the highest scales. They satisfy valence and momentum sum rules at the percent level and reproduce representative pp toπ and pp toγπ cross sections within approximately 2% and 5%, respectively, in the kinematic regions studied. Compared with grid interpolation through LHAPDF, direct evaluation of the analytic expressions reduces the total Monte Carlo run time by about 11% and 30% in these two applications.
Sources
- Combination of Measurements of Inclusive Deep Inelastic $e^{\pm}p$ Scattering Cross Sections and QCD Analysis of HERA Data
- LHAPDF6: parton density access in the LHC precision era
- Using analytic models to describe effective PDFs
- Hadron plus photon production in polarized hadronic collisions at next-to-leading order accuracy
- Analysis of the internal structure of hadrons using direct photon production
- A quantum representation of pi fragmentation functions through variational quantum circuits
Related papers
- MadNIS at NLO
- The NANOGrav 15-year data set: A Limit on Continuous Spin Gravitons
- DIS-secting the universe with high-energy neutrinos
- Crowdsourcing Gravitational Waves from Superradiant Axions
- Classification of g-modes for neutron stars with a strong transition: Novel universal relation including slow stable hybrid stars
- Scalar axion field of toroidal electromagnetic pulses