Inclusion of the $^3P_0$ model in PYTHIA 8

Albi Kerbizi; Leif L\"onnblad

REVIEW 3 major objections 5 minor 10 references

Inclusion of the $^3P_0$ model in PYTHIA 8

T0 review · 3 major / 5 minor · reviewed 2026-08-14 · deepseek-v4-flash

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

Pith's one-line read This paper integrates the $^3P_0$ model of polarized quark fragmentation into the PYTHIA 8 event generator, enabling spin effects in hadronization for the first time, and shows that the resulting Collins and di-hadron asymmetries follow…

desk verdict Genuinely new: spin-dependent hadronization in PYTHIA 8, but the COMPASS agreement is inherited from inputs, not yet an independent test. read the letter →

arxiv 1909.02280 v1 pith:IV4AOS7F submitted 2019-09-05 hep-ph

Albi Kerbizi , Leif Lönnblad This is my paper · ORCID

classification hep-ph PACS 13.88.+e

keywords spineffectshadronizationPYTHIA83P0modelCollinsasymmetrydi-hadrontransversitySIDIS

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 reports the first implementation of spin effects in the hadronization stage of the PYTHIA 8 event generator. The spin dependence is carried by the $^3P_0$ model of string breakup, which adds one complex mass parameter to the standard Lund mechanics and propagates the fragmenting quark's polarization along the hadronization chain. The authors simulate transversely polarized SIDIS at COMPASS kinematics and find that the simulated Collins and di-hadron asymmetries follow the same trends as the COMPASS data, with slightly larger amplitudes. They also show that the interface leaves the unpolarized $z_h$ and $p_T^2$ distributions essentially unchanged, and they identify the neglect of vector-meson decays and retuning of the model parameter as the next steps.

What carries the argument

The carrying mechanism is the $^3P_0$ string-breakup model: at each break a $q\bar q$ pair is created with vacuum quantum numbers, and the hadron formed from the fragmenting quark and the new antiquark is accepted with probability $$w(k'_T,S'_q)=\frac12\left(1-\frac{2\,\operatorname{Im}(\mu)\, S'_q\cdot(\hat z\times k'_T)}{|\mu|^2+k'^2_T}\right),$$ where $\mu=(0.42+i0.76)$ GeV is the single complex mass parameter of the model, $k'_T$ is the transverse momentum of the produced antiquark, and $S'_q$ is the polarization vector of the fragmenting quark. This weight turns quark transverse polarization into a left-right (Collins) asymmetry of the produced hadrons. After each accepted break the polarization of the produced quark is updated by the 3P0 spin-propagation rules, so spin information travels down the string. The interface preserves PYTHIA's standard exit procedure for the last two hadrons of each string, which are produced without spin effects.

What would settle it

Trace how the value of $\mu=(0.42+i0.76)$ GeV was determined in the standalone 3P0 Monte Carlo, or refit $\operatorname{Im}(\mu)$ to an independent Collins observable such as the $e^+e^-$ asymmetry; if the value required to reproduce COMPASS differs from the one used here, the claimed agreement is a consistency check, not a prediction.

Watch

Extended reading notes

Core claim

The central claim is that the $^3P_0$ model of polarized quark fragmentation can be embedded in PYTHIA 8 without disturbing the generator's unpolarized hadron distributions, and that the resulting PYTHIA+3P0 simulation of transversely polarized SIDIS at COMPASS kinematics produces Collins and di-hadron asymmetries whose amplitude and kinematic dependence follow the same trends as the COMPASS measurements, with a slightly larger amplitude that the authors attribute to the neglect of vector-meson decays and to the need to retune the imaginary part of the model parameter $\mu$. For the first time in this generator, the quark's transverse polarization is propagated through the string breakups rather than discarded.

Load-bearing premise

The agreement with COMPASS data rests on the assumption that the model's one free spin parameter was fixed without using the data it is compared with; the paper gives no evidence that it was.

Editorial extensions

If this is right

The same PYTHIA+3P0 code can produce multidimensional Collins and di-hadron asymmetry maps in $x_B$, $z_h$, and $p_T$ for proton and neutron targets without changing the underlying hadronization tune.
The simulation reproduces the near-zero deuteron Collins and di-hadron asymmetries observed by COMPASS, which follows from the cancellation between proton and neutron contributions in the implementation.
The interface leaves standard PYTHIA momentum distributions essentially unchanged, so spin effects can be turned on or off without breaking existing event-generation workflows.
Because the current amplitudes are slightly too large, the paper identifies retuning of $\operatorname{Im}(\mu)$ and the eventual inclusion of vector-meson decays as the immediate next steps.

Reading between the lines

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

Editorial inference: the model's single free parameter $\operatorname{Im}(\mu)$ is a natural target for a global fit combining COMPASS Collins data with $e^+e^-$ Collins measurements; a value fixed by the latter would turn the present consistency check into a genuine prediction.
Editorial inference: because the spin-propagation logic is attached to individual string breaks, the same interface should apply to other hard processes, not just SIDIS; polarized fragmentation in $e^+e^-$ annihilation is the nearest place to test it.
Editorial inference: the paper's intermediate-$z_h$ discrepancy points to vector-meson production as the next missing piece; including polarized vector-meson decays would show whether the 3P0 weight alone accounts for the Collins effect.

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.

Desk Editor's Note

read the letter

Albi and Leif's proceedings note does something genuinely new: it puts the 3P0 spin-dependent fragmentation model inside PYTHIA 8, which no one had done before. The implementation path is sensible — an accept/reject weight for a single complex mass parameter, followed by propagation of the quark polarization along the string — and the validation against their own standalone 3P0 MC is the right check for an interface. The fact that zh and pT distributions survive the interface, and that the simulated Collins and di-hadron asymmetries match the standalone code, tells me the spin machinery is not being distorted by PYTHIA's string handling. They also state their approximations honestly: only pseudoscalars, a spinless final two-hadron exit, and three-hadron strings left out.

The soft spot is the COMPASS comparison. The paper calls it 'very satisfactory,' but that agreement is probably inherited. The complex mass mu = (0.42 + i0.76) GeV comes from their earlier standalone model [4], and the paper never says how that value was fixed. If Im(mu) was chosen to reproduce the same Collins or di-hadron data, then Figs. 2-3 are a consistency check, not an independent test. The same goes for the transversity parametrization from Ref. [10], which is a phenomenological extraction that very likely used COMPASS Collins data. So the empirical validation is weaker than the prose suggests. On top of that, the simulated points have no error bars, and no code or data artifact is released, so a reader cannot reproduce the numbers today.

None of this kills the central claim. The interface itself is the deliverable, and the internal validation supports it. But the COMPASS agreement should be described as qualitative and preliminary, with the parameter-fitting history disclosed. The fix is straightforward: state where mu and the transversity inputs come from, add uncertainties to the MC points, and put the interface on a public repository. This is a proceedings paper, so I would treat it as a status report rather than a final result. It deserves a serious referee if submitted to a journal — the implementation is important enough — but it would need those revisions before acceptance. For a reading group, it's worth a look for people who care about spin effects in event generators.

Referee Report

3 major / 5 minor

Summary. The manuscript reports the first implementation of spin effects in the hadronization step of the PYTHIA 8 event generator, based on the $^3P_0$ model of polarized quark fragmentation. The authors describe an interface in which each string breakup is accepted with a spin-dependent weight Eq. (2.1) governed by a complex mass parameter $\mu$, and the quark polarization is propagated along the fragmentation chain, limited to pseudoscalar mesons and with the final two hadrons produced without spin weight. The implementation is validated by comparing transverse-momentum and energy-fraction distributions and Collins/di-hadron asymmetries with the standalone $^3P_0$ Monte Carlo, showing agreement. The paper then presents preliminary simulations of transversely polarized SIDIS at COMPASS kinematics, comparing Collins and di-hadron asymmetries for proton and deuteron targets with COMPASS data; the simulated asymmetries reproduce the qualitative trends but are somewhat larger in magnitude, and the authors state that retuning of $\mathrm{Im}(\mu)$ would be needed.

Significance. If the implementation is correct, this work offers a practical tool for generating spin-dependent hadronization observables within a full event generator, enabling multidimensional studies and future predictions. The technical validation against the standalone $^3P_0$ Monte Carlo is the appropriate check for an interface and is convincing as far as it goes. The paper also states its principal limitations clearly: only pseudoscalar mesons are produced, the last two hadrons in the fragmentation chain are unweighted, and vector-meson decays and certain three-hadron strings are neglected. The main weakness is that the empirical validation against COMPASS data in Figs. 2 and 3 is potentially circular: the complex mass $\mu$ is imported from Ref. [4] and the transversity parametrizations from Ref. [10] without any disclosure of whether those inputs were tuned to the same COMPASS data used for comparison.

major comments (3)

[Section 3, Eq. (2.1), Figs. 2-3] The paper does not state how the complex mass parameter $\mu = (0.42 + i\,0.76)$ GeV was fixed in the standalone $^3P_0$ Monte Carlo of Ref. [4]. Since $\mathrm{Im}(\mu)$ controls the size of the Collins asymmetry, and the text admits that retuning $\mathrm{Im}(\mu)$ is needed, the comparison with COMPASS data in Figs. 2 and 3 cannot be interpreted as an independent validation unless the fitting history of $\mu$ is disclosed. If $\mu$ was adjusted using the same COMPASS data shown here, then the observed agreement is a consistency check, not a new prediction. Please provide the provenance of $\mu$ with respect to the compared data and temper the claim that the comparison is "very satisfactory" accordingly.
[Section 3, transversity parametrizations] The transverse polarization of the fragmenting quark is set using the parametrizations $xh_u^1 = 3.2\,x^{1.28}(1-x)^4$ and $xh_d^1 = -4.6\,x^{1.44}(1-x)^4$ taken from Ref. [10]. If these parametrizations were extracted from the same COMPASS Collins or di-hadron asymmetries used for comparison in Figs. 2-3, then those figures again do not constitute an independent test of the model. The manuscript should clarify whether Ref. [10] used the COMPASS data shown here, and if so, the conclusions should be rephrased from validation to a consistency check.
[Section 3, exit condition] The paper states that the last two hadrons completing the fragmentation are produced without spin effects and that strings fragmenting into three hadrons where only one is weighted are neglected. Since the di-hadron asymmetry in Fig. 3 is by construction built from pairs of hadrons in the same jet, the final two hadrons in the chain are exactly the type of pair that enters this observable. The possible bias introduced by the unweighted exit condition and by the neglected three-hadron strings is not quantified. Please provide an estimate of the fraction of pairs affected by these approximations or a cross-check with an alternative treatment of the final two hadrons to justify that the comparison with COMPASS in Fig. 3 is robust.

minor comments (5)

[References] Reference [2] combines two distinct publications (JHEP05 (2006) 026 and Comput.Phys.Commun. 191 (2015) 159-177) into a single citation; these should be split into separate entries.
[Section 4] The word "multi-dimentional" should be "multi-dimensional".
[Section 2] The sentence "PYTHIA starts the hadronization by generating a first break-up with aq′ ¯q′ pair" contains a spacing error; it should read "with a $q'\bar{q}'$ pair".
[Section 3, Fig. 2] The label $a_{u^\uparrow\to h+X}$ for the Collins analysing power is not defined in the text; please introduce the notation or use the standard $A_{UT}^{\sin(\phi_C)}$ convention used in the COMPASS figures.
[Figures 2-3] Several axes and labels in the figures are too small or partially illegible; please provide higher-resolution figures or larger fonts.

Circularity Check

1 steps flagged · score 4.0 of 10

The COMPASS 'agreement' is inherited from the authors' own fitted 3P0 parameter; the PYTHIA implementation itself is independently checked.

fitted input called prediction [Section 3, Eq. (2.1), Figs. 2–3]
"We accept the hadron according to the weight w(k′T,S′q)= 1/2 (1 − 2Im(µ)S′q·(ˆz×k′T)/(|µ|2 + k′2T)), (2.1) obtained from the splitting function of the 3P0 model in Ref. [4]. ... The complex mass has been taken µ = (0.42 + i0.76) GeV [4]."

Eq. (2.1) is the only spin-dependent acceptance weight; it generates exactly the Collins and di-hadron modulations shown in Figs. 2–3, with amplitudes controlled by Im(µ). This parameter is not determined in the present paper or tied to an independent observable; it is imported from Ref. [4], the authors' own standalone 3P0 Monte Carlo. The paper supplies no fitting history or constraint showing that Im(µ) was not fixed using the same COMPASS spin-asymmetry data used for the 'very satisfactory' comparison. The claimed empirical validation is therefore at best a consistency check of an inherited fitted value, not an independent prediction of PYTHIA+3P0.

full rationale

The software-inclusion claim is self-contained: the comparison of PYTHIA+3P0 with the standalone 3P0 MC in Fig. 1 and the matching Collins analysing power and di-hadron asymmetries validate that the spin weight Eq. (2.1) was correctly transferred into PYTHIA, so the central technical implementation is not circular. The circularity concern is confined to the empirical benchmarking. The spin observables are produced by Eq. (2.1), whose only spin-scale parameter is Im(µ), and the paper takes µ from Ref. [4] without showing that this value is independent of the COMPASS asymmetries used for comparison. If Im(µ) was tuned in that prior work to the same kind of Collins or di-hadron data, the 'very satisfactory' agreement in Figs. 2–3 is inherited rather than predicted. The transversity parametrizations of Ref. [10] similarly enter the simulation as an input extracted from the same family of SIDIS data. Because the implementation is independently verified while the empirical validation rests on inputs whose fitting history is not disclosed, the paper is partially, but not wholly, circular.

Assumptions & free parameters 6 free parameters · 5 assumptions · 0 invented entities

No new particles, forces, or conserved quantities are introduced. The 3P0 q-qbar break-up state is a model assumption, not a new entity. The central claim depends on several prior fitted parameters, especially the complex mass mu, and on domain assumptions about the Lund model, transversity PDFs, and neglected remnant polarization.

free parameters (6)

Complex 3P0 mass mu = (0.42 + i 0.76) GeV
Controls the spin-dependent weight in Eq. (2.1); taken from the authors' standalone model [4] rather than derived or fitted in this paper.
Lund a parameter = 0.9
String fragmentation parameter, set to the value used in the standalone 3P0 Monte Carlo [4].
Lund b parameter = 0.5 GeV^-2
String fragmentation parameter, set to the value used in the standalone 3P0 Monte Carlo [4].
Intrinsic kT width b_T^-1/2 = 0.34 GeV
Gaussian width of intrinsic transverse momentum, set to the standalone 3P0 Monte Carlo value [4].
u-quark transversity parameterization = x h_u1 = 3.2 x^1.28 (1-x)^4
Taken from Ref. [10] and used in Section 3 to set the transverse polarization of u quarks before hard scattering.
d-quark transversity parameterization = x h_d1 = -4.6 x^1.44 (1-x)^4
Taken from Ref. [10] and used in Section 3 to set the transverse polarization of d quarks before hard scattering.

assumptions (5)

domain assumption Quark-antiquark pairs at string breakups are produced with vacuum quantum numbers L=0, S=1, J=0 as in the 3P0 model.
This is the defining postulate of the 3P0 model, invoked through Eq. (2.1) and the spin-transfer rules of Ref. [4].
domain assumption The Lund string fragmentation model correctly describes unpolarized hadronization.
PYTHIA 8's string hadronization is the base that the 3P0 model modifies; standard in the field but unproved in this paper.
domain assumption The transversity PDF parametrizations of Ref. [10] are valid at the COMPASS kinematics used here.
Used in Section 3, Eq. (3.1), to set the initial quark polarization before hard scattering.
domain assumption The polarization of the target remnant can be neglected.
Stated in Section 2: the string is set up between the active quark and the diquark remnant with no remnant polarization.
ad hoc to paper Completing hadronization with the final two hadrons unweighted does not bias the spin asymmetries.
The interface lets PYTHIA's standard exit procedure produce the last two hadrons without 3P0 spin weights, an approximation introduced in Section 2.

how reviews work

0 comments

Cite this review

Pith. "Pith review of Inclusion of the $^3P_0$ model in PYTHIA 8." pith.science (2026). https://pith.science/paper/IV4AOS7F

@misc{pith2026190902280,
  author       = {Pith},
  title        = {Pith review of: Inclusion of the $^3P_0$ model in PYTHIA 8},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/IV4AOS7F}},
  note         = {Machine review of arXiv:1909.02280}
}

abstract

The spin effects in the hadronization process have been included for the first time in the PYTHIA 8 event generator. The spin effects are limited to the production of pseudo-scalar mesons and are obtained from the propagation of the quark polarization along the fragmentation chain according to the rules of the so-called $^3P_0$ model. The interface between PYTHIA 8 and the package of the $^3P_0$ model is presented together with preliminary results on the Collins and di-hadron asymmetries as obtained from simulations of the transversely polarized semi-inclusive deep inelastic scattering process.

Figures

Figures reproduced from arXiv: 1909.02280 by the authors.

**Figure 2.** Collins proton asymmetry as function of xB, zh and pT for positive pions (circles) and negative pions (triangles) as obtained from PYTHIA+3P0 (upper row) compared to COMPASS data [6] (lower row). The corresponding results have been produced also for a deuteron target by merging simulations performed separately for a proton and for a neutron target. From simulations we obtain very small values (below 1%) for the Col… view at source ↗

**Figure 3.** Di-hadron asymmetry for pairs of oppositely charged hadrons in the same jet, as func [PITH_FULL_IMAGE:figures/full_fig_p006_3.png] view at source ↗

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

10 extracted references · 9 canonical work pages

[4]

Kerbizi, X

A. Kerbizi, X. Artru, Z. Belghobsi and A. Martin, arXiv:1903.01736 [hep-ph] (accepted for publication in Phys. Rev. D)

arXiv 1903
[10]

Martin, F

A. Martin, F. Bradamante and V . Barone, Phys. Rev. D91 (2015) no.1, 014034. 5

work page 2015
[1]

J. C. Collins, Nucl. Phys. B 396 (1993) 161

work page 1993
[2]

Sjöstrand, S

T. Sjöstrand, S. Mrenna and P. Skands, JHEP05 (2006) 026, Comput.Phys.Commun. 191 (2015) 159-177

work page 2006
[3]

Kerbizi, X

A. Kerbizi, X. Artru, Z. Belghobsi, F. Bradamante and A. Martin, Phys. Rev. D 97 (2018) no.7, 074010

work page 2018
[5]

Andersson, G

B. Andersson, G. Gustafson, G. Ingelman and T. Sjostrand, Phys. Rept. 97 (1983) 31

work page 1983
[6]

Adolph et al

C. Adolph et al. [COMPASS Collaboration], Phys. Lett. B 744 (2015) 250

work page 2015
[7]

Adolph et al

C. Adolph et al. [COMPASS Collaboration], Phys. Lett. B 736 (2014) 124

work page 2014

Show all 10 references

[8]

Alekseev et al

M. Alekseev et al. [COMPASS Collaboration], Phys. Lett. B 673 (2009) 127

2009
[9]

Adolph et al

C. Adolph et al. [COMPASS Collaboration], Phys. Lett. B 713 (2012) 10

2012

Pith tools

Reviewed August 14, 2026 · model on record in the stance chip above.

[1] [4]

Kerbizi, X

A. Kerbizi, X. Artru, Z. Belghobsi and A. Martin, arXiv:1903.01736 [hep-ph] (accepted for publication in Phys. Rev. D)

arXiv 1903

[2] [10]

Martin, F

A. Martin, F. Bradamante and V . Barone, Phys. Rev. D91 (2015) no.1, 014034. 5

work page 2015

[3] [1]

J. C. Collins, Nucl. Phys. B 396 (1993) 161

work page 1993

[4] [2]

Sjöstrand, S

T. Sjöstrand, S. Mrenna and P. Skands, JHEP05 (2006) 026, Comput.Phys.Commun. 191 (2015) 159-177

work page 2006

[5] [3]

Kerbizi, X

A. Kerbizi, X. Artru, Z. Belghobsi, F. Bradamante and A. Martin, Phys. Rev. D 97 (2018) no.7, 074010

work page 2018

[6] [5]

Andersson, G

B. Andersson, G. Gustafson, G. Ingelman and T. Sjostrand, Phys. Rept. 97 (1983) 31

work page 1983

[7] [6]

Adolph et al

C. Adolph et al. [COMPASS Collaboration], Phys. Lett. B 744 (2015) 250

work page 2015

[8] [7]

Adolph et al

C. Adolph et al. [COMPASS Collaboration], Phys. Lett. B 736 (2014) 124

work page 2014