Building Next-to-Next Leading Logarithmic parton showers: the PanScales recipe

Silvia Ferrario Ravasio

REVIEW 4 major objections 6 minor 1 cited by

Building Next-to-Next Leading Logarithmic parton showers: the PanScales recipe

T0 review · 4 major / 6 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 paper claims a recipe that produces the first parton showers with demonstrable next-to-next-to-leading-logarithmic (NNLL) accuracy for global event shapes and for soft-sensitive non-global observables in e+e− collisions.

desk verdict A clean, readable proceedings summary of a genuine PanScales milestone, but the NNLL claim is stated more boldly than the text can support; all load-bearing derivations live in the cited PRLs. read the letter →

arxiv 2505.13395 v1 pith:SHNL2UNU submitted 2025-05-19 hep-ph

Silvia Ferrario Ravasio This is my paper · ORCID

classification hep-ph

keywords partonshowersNNLLresummationglobaleventshapesnon-globalobservablesLundplanee+e−collisionsQCDMonteCarlogenerators

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

Parton showers are the default tool for turning QCD predictions into fully exclusive collider-event descriptions, but the showers used at the LHC have long been limited to leading or next-to-leading logarithmic accuracy. This paper's central claim is that a specific combination of corrections—NLO matching, a double-soft emission correction, the universal constants $K_1$ and $K_2$, and drift-compensation factors—yields the first parton showers with next-to-next-to-leading-logarithmic (NNLL) accuracy for any global event shape in e+e− collisions, and for non-global observables that are primarily sensitive to soft emissions. If the claim holds, these showers close much of the formal-accuracy gap between analytic resummation and Monte Carlo event generators, giving collider phenomenology an exclusive event generator with NNLL logarithmic accuracy. The paper also states that genuine triple-collinear dynamics, while not needed for these observables, remains necessary for general NNLL accuracy.

What carries the argument

The Lund plane—a map of radiative phase space with $\ln k_t$ and rapidity $y$ as coordinates—is the central bookkeeping device: it sorts emissions into hard-collinear, soft large-angle, and soft-collinear regions and tracks how later branchings drift earlier emissions. The other load-bearing object is the Monte Carlo scheme for the strong coupling, $\alpha_s^{\mathrm{MC}}(k_t)=\alpha_s(k_t)[1+(\alpha_s/2\pi)K_1]$, which encodes the universal single-emission constant. Around these, the recipe adds ratio corrections from the exact double-soft matrix element, the NNLO constant $K_2$, the hard-collinear average $B_2(z)$, and drift factors $\Delta K_1$, $\Delta\ln k_t$, and $\Delta\ln z$ that compensate for coordinate shifts introduced by the shower's momentum mapping.

What would settle it

A direct check would compare this shower's thrust distribution to a complete analytic next-to-next-to-leading-log resummation; a residual difference at relative order $\alpha_s^2$ with logarithmic enhancement beyond the included terms would falsify the completeness claim. More narrowly, a fixed-order calculation demonstrating that triple-collinear splitting contributes to the NNLL thrust coefficient would directly refute the assertion that these observables are NNLL without triple-collinear dynamics.

Watch

Extended reading notes

Core claim

The discovery is a constructive recipe: start from an NLL-accurate dipole shower, then add (i) NLO matching to capture the $\alpha_s(\alpha_s L)^n$ terms, (ii) a correction to the double-soft emission rate when two soft partons are close in the Lund plane, (iii) the universal constants $K_1$ and $K_2$ in a properly defined Monte Carlo $\alpha_s$ scheme with three-loop running, (iv) drift-compensation factors $\Delta K_1(y)$, $\Delta\ln k_t$, and $\Delta\ln z$ that correct for the way the shower's kinematic map shifts emissions in rapidity and transverse momentum, and (v) the average hard-collinear correction $B_2(z)$. The paper asserts that this combination is sufficient to make the shower NNLL-accurate for any global event shape, and for non-global observables mainly sensitive to soft emissions; it further reports improved agreement with e+e− event-shape data once these corrections are included.

Load-bearing premise

The load-bearing premise is that, for global event shapes, the listed corrections are complete at this order of log accuracy, so genuine three-parton collinear splitting dynamics is not needed for those observables.

Editorial extensions

If this is right

If the recipe is correct, e+e− event-shape predictions from these showers carry NNLL logarithmic accuracy, meaning their logarithmic uncertainty is of order $\alpha_s^2(\alpha_s L)^n$ rather than $\alpha_s(\alpha_s L)^n$.
For non-global observables dominated by soft emissions, the shower achieves the same accuracy as state-of-the-art analytic resummations, not just a formal claim but a demonstrable one.
The corrections improve agreement with LEP event-shape data for thrust and the Durham jet-resolution variable $y_{23}$, as shown in the paper's comparison plots.
The paper leaves correct triple-collinear dynamics and extensions to incoming partons and multiple hard jets as the next milestones, implying the present accuracy does not yet cover all hadron-collider processes.

Reading between the lines

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

One could view the drift-compensation factors as a general design rule: any shower whose recoil strategy shifts an emission's resolved coordinates must locally recompute the scheme-dependent constants, a principle likely to transfer to hadron-collider showers.
A testable extension is to apply the same recipe to further global event shapes, such as C-parameter or heavy-jet mass, and compare against analytic NNLL resummations; flat ratios would corroborate the completeness assumption.
If the completeness assumption is later found to fail for some global observable, the likely culprit is a triple-collinear term, making the explicit omission of triple-collinear dynamics the clearest target for scrutiny.

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

Bottom line: this is a good summary of an important result, but it is a summary, not a primary source. The first NNLL-accurate PanScales showers were published in the cited PRLs, and this text adds no new derivation, data, or algorithm. If you want the proof, you go to Refs. 1 and 16.

What the paper does well: it collects the recipe in one place and explains it clearly. The Lund-plane picture is a nice organizing device, and the list of ingredients—NLO matching for the hard region, double-soft acceptance, ΔK1, K2, the drift corrections, and the averaged B2—is presented in a way that a non-specialist can follow. The comparison to ALEPH data in Fig. 1b is also useful: the NNLL shower visibly improves on the NLL one, which is credible evidence that the construction is behaving as intended.

The soft spots are real but not fatal. Section 4 claims NNLL accuracy for "any global event shape" without proving that global event shapes are outside the class of observables that need triple-collinear dynamics. The paper itself says triple-collinear dynamics is necessary for general NNLL accuracy, so the line between "global" and "general" is doing a lot of work, and no argument for the exclusion is given here. Similarly, the averaging of B2(z) over z is stated without justification for why an unweighted average suffices for any global shape. These are legitimate gaps, and the stress-test note lands on the right spot. That said, this is a proceedings; the missing derivations are in the cited PRLs, so the gaps are limitations of the format rather than red flags about the underlying physics.

The citation pattern is appropriate: the collaboration cites its own published papers for the technical steps, which is exactly what a proceedings should do. There is no external verification in this text, but the PRLs provide the support.

Who gains from this: anyone who wants a quick map of the PanScales strategy without digging through three PRLs, or someone teaching parton showers who needs a compact overview. It is not a new result and should not be cited as one. I would send it to peer review, but with a request to qualify the abstract and Section 4 statements—add a sentence saying the proof is in Refs. 1 and 16, and soften "any global event shape" to something like "the classes of global event shapes studied in Refs. 1 and 16." The underlying work is solid, so this is a correction in presentation, not a rejection.

Referee Report

4 major / 6 minor

Summary. This proceedings paper summarizes the PanScales collaboration's recipe for constructing parton showers with claimed next-to-next-to-leading-logarithmic (NNLL) accuracy in e+e- collisions. The recipe consists of four ingredients: NLO matching for the hard region, double-soft matrix-element corrections together with a drift correction Delta K1 for the soft region, the two-loop coupling scheme with K2 and drift corrections for the soft-collinear region, and an averaged NLO correction B2(z) for the hard-collinear region. The paper claims that this combination yields the first parton shower with NNLL accuracy for any global event shape and for non-global observables primarily sensitive to soft emissions, and it shows a comparison with ALEPH data for thrust and the Durham jet resolution y23. The derivations of the individual ingredients and the numerical logarithmic-accuracy tests are delegated to the collaboration's earlier papers, in particular Refs. 1, 16, and 25.

Significance. If the central claim is correct, this is an important milestone: the first parton showers with demonstrable NNLL accuracy for broad classes of e+e- observables, connecting analytic resummation with general-purpose Monte Carlo simulations. The paper is useful as a high-level statement of the PanScales strategy and is honest about an important limitation, namely that triple-collinear dynamics is still needed for 'general NNLL accuracy.' However, this proceedings does not itself demonstrate the central claim: the technical derivations and accuracy tests are not included, and the data comparison in Fig. 1b is an illustration of improved agreement rather than a quantitative log-accuracy test. The significance for the field is therefore real but conditional on the cited companion papers.

major comments (4)

[Section 4 (Concluding remarks)] The central claim that the combination of ingredients yields the first NNLL-accurate parton shower for any global event shape is stated without demonstration in this manuscript. The derivations of K1, Delta K1, the double-soft acceptance probability, B2, and the logarithmic-accuracy tests are all delegated to Refs. 1, 16, and 25. Figure 1b shows improved agreement with ALEPH data but is not a quantitative test of logarithmic accuracy. As written, the paper does not provide evidence for the word 'demonstrably' in the abstract. Please present a brief accuracy test, such as a comparison with analytic NNLL predictions for logarithmic moments, or explicitly reframe the claim as reporting results established in the cited papers.
[Section 4 (Concluding remarks) and Section 3 (hard-collinear region)] The manuscript asserts that global event shapes already achieve NNLL accuracy despite the statement that 'the attainment of the correct triple-collinear dynamics' is 'necessary for general NNLL accuracy.' No argument is given to show that global event shapes do not receive NNLL-level triple-collinear contributions. This distinction is load-bearing: if a triple-collinear term contributes to the NNLL coefficient of, say, thrust or C-parameter, the central claim fails. Please provide the explicit argument or cite the precise result in Refs. 1 or 16 that establishes the exclusion for global event shapes.
[Section 3 (NNLL corrections to global event shapes from the hard-collinear region)] The 'average NLO correction B2(z)' is not defined: it is unclear whether the z-integral is weighted by the observable's measurement function or is an unweighted average over the full z range. For a general global event shape, the hard-collinear contribution to the NNLL coefficient can depend on how the observable responds to the splitting fraction z, so an unweighted average may not be shape-independent. Please state the definition of the average and justify its universality, or restrict the claim to observables for which the averaging is valid.
[Section 3 (soft and soft-collinear regions)] Several central ingredients of the recipe are introduced only verbally: the drift corrections Delta K1(y), Delta ln k_t,sc, and Delta ln z_sc appear in the text without definitions or explicit formulas. Since these corrections are part of the claimed NNLL construction, the reader cannot verify the recipe from this manuscript alone. Please provide the definitions or give the specific equation numbers in Refs. 1 and 16 where they are defined.

minor comments (6)

[Section 1] The phrase 'PanScalescollaboration' is missing a space and should read 'PanScales collaboration'.
[Equation (1)] The typesetting of Eq. (1) is inconsistent: 'alpha_MC^s' and 'alpha_s(k_t)' use different spacing and subscript/superscript conventions. Please unify the notation.
[References] References 16 and 28 have incomplete article numbers, reading '16' and '8' respectively; please complete the bibliographic information.
[Figure 1b] The legend labels 'PGsdf 0 PG0 PG1/2' are unclear; please identify which curve corresponds to which shower variant in a way that is legible in print.
[Section 1 (citations)] The citation order in the sentence listing NLL showers, [9,10,11,13,12,14,15], is not sequential; please reorder the references.
[Abstract and Section 4] The abstract claims 'first demonstrably NNLL-accurate parton showers' without specifying the collision type, while Section 4 states the claim for e+e- collisions. Please harmonize the scope of the claim.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity; the recipe's ingredients are fixed by analytic QCD matrix elements and checked against external ALEPH data.

full rationale

This proceedings summarizes a construction strategy; it does not attempt to derive NNLL accuracy from scratch. The load-bearing ingredients are NLO matching (Ref. 25), the double-soft matrix-element ratio and DeltaK1 (Ref. 16), the K2 and drift corrections (Refs. 18,1), and the B2(z) hard-collinear correction (Refs. 19,20). These coefficients are fixed by analytic resummation and fixed-order matrix-element calculations; they are not fitted to the shower's own NNLL claims. The final comparison to ALEPH data (Fig. 1b) provides an external falsification channel independent of the author-derived resummation scheme. The Section 4 statement that triple-collinear dynamics is necessary for 'general NNLL accuracy' while global event shapes are already NNLL is an unproven completeness assumption; that is a correctness risk rather than circularity, because the paper does not define global-event-shape NNLL in terms of the absence of triple-collinear terms. No equation is equivalent to its inputs by construction, and no fitted parameter is renamed as a prediction. The appropriate finding is therefore no circularity.

Assumptions & free parameters 1 free parameters · 4 assumptions · 0 invented entities

The paper is a proceedings summary, so the physics input is pulled from cited analytic resummations and the collaboration's own PRLs. The assumptions a reader must accept are planar colour, factorisation by colour coherence, universality of the imported constants, and completeness of the ingredient list for global event shapes. One free parameter, the shower cutoff, is mentioned but not expected to affect NNLL predictions. No new entities are invented.

free parameters (1)

Non-perturbative cutoff Lambda = not specified
The dipole shower stops emissions below the cutoff Lambda; NNLL predictions are expected to be insensitive to its value, but it is a hand-chosen parameter of the algorithm.

assumptions (4)

domain assumption Large-number-of-colours (planar) limit for colour-ordered dipoles.
Invoked in Section 2 to define a dipole as a pair of colour-connected partons; subleading-colour effects are neglected.
domain assumption Colour coherence implies the matrix element factorises into a product of independent emissions for multiple soft-collinear emissions.
Used in Sections 2 and 3 to justify treating emissions as independent and to build the Lund-plane picture.
domain assumption The analytic resummation constants K1, K2, B2 and the double-soft matrix element are universal and transferable to the parton shower phase space.
The construction in Section 3 imports these ingredients from Refs 17, 18, 19, 20 and Ref 16 without re-deriving them.
domain assumption For global event shapes, the listed NNLL corrections are exhaustive and full triple-collinear dynamics is not required.
Section 4 states triple-collinear dynamics is necessary for general NNLL accuracy but claims global event shapes are already NNLL without it; this completeness assumption is the paper's weakest premise.

how reviews work

0 comments

Cite this review

Pith. "Pith review of Building Next-to-Next Leading Logarithmic parton showers: the PanScales recipe." pith.science (2026). https://pith.science/paper/SHNL2UNU

@misc{pith2026250513395,
  author       = {Pith},
  title        = {Pith review of: Building Next-to-Next Leading Logarithmic parton showers: the PanScales recipe},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/SHNL2UNU}},
  note         = {Machine review of arXiv:2505.13395}
}

read the original abstract

Parton showers lie at the core of Shower Monte Carlo event generators, the default theoretical tools used to interpret collider data. In these proceedings, we summarise the strategy of the PanScales collaboration that led to the attainment of the first demonstrably next-to-next-to-leading-logarithmic accurate parton showers.

Discussion (0). Continue with ORCID to comment.

Forward citations

Cited by 1 Pith paper

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

Theoretical Summary: Moriond QCD and High-Energy Interactions 2025
hep-ph 2025-06 unverdicted

A conference summary that compiles the main theory results presented at Moriond QCD 2025, spanning hard scattering, precision QCD, flavour, strong coupling, lattice, heavy-ion, and BSM physics.

Reference graph

Works this paper leans on

29 extracted references · 28 canonical work pages · cited by 1 Pith paper

[1]

van Beekveld, M

M. van Beekveld, M. Dasgupta,et al.Phys. Rev. Lett.134(2025) no.1, 011901

work page 2025
[2]

C. Duhr, B. Mistlberger and G. Vita, Phys. Rev. Lett.129(2022) no.16, 162001

work page 2022
[3]

X. Chen, T. Gehrmann,et al.Phys. Rev. Lett.128(2022) no.25, 252001

work page 2022
[4]

Camarda, L

S. Camarda, L. Cieri and G. Ferrera, Phys. Lett. B845(2023), 138125

work page 2023
[5]

M. Bahr, S. Gieseke, M. A. Gigg,et al.[Herwig], Eur. Phys. J. C58(2008), 639-707

work page 2008
[6]

Bierlich, S

C. Bierlich, S. Chakraborty,et al.[Pythia], SciPost Phys. Codeb.2022(2022), 8

work page 2022
[7]

Bothmannet al.[Sherpa], JHEP12(2024), 156

E. Bothmannet al.[Sherpa], JHEP12(2024), 156

work page 2024
[8]

Dasgupta, F

M. Dasgupta, F. A. Dreyer,et al.JHEP09(2018), 033 [erratum: JHEP03(2020), 083]

work page 2018

Show all 29 references

[9]

Dasgupta, F

M. Dasgupta, F. A. Dreyer, K. Hamilton,et al.Phys. Rev. Lett.125(2020) no.5, 052002

2020
[10]

Nagy and D

Z. Nagy and D. E. Soper, Phys. Rev. D104(2021) no.5, 054049

2021
[11]

J. R. Forshaw, J. Holguin and S. Pl¨ atzer, JHEP09(2020), 014

2020
[12]

van Beekveld, S

M. van Beekveld, S. Ferrario Ravasio, G. P. Salam,et al.JHEP11(2022), 019

2022
[13]

Herren, S

F. Herren, S. H¨ oche, F. Krauss, D. Reichelt and M. Schoenherr, JHEP10(2023), 091

2023
[14]

van Beekveld and S

M. van Beekveld and S. Ferrario Ravasio, JHEP02(2024), 001

2024
[15]

C. T. Preuss, JHEP07(2024), 161

2024
[16]

Ferrario Ravasio, K

S. Ferrario Ravasio, K. Hamilton,et al.Phys. Rev. Lett.131(2023) no.16, 16

2023
[17]

Catani, B

S. Catani, B. R. Webber and G. Marchesini, Nucl. Phys. B349(1991), 635-654

1991
[18]

Banfi, B

A. Banfi, B. K. El-Menoufi and P. F. Monni, JHEP01(2019), 083

2019
[19]

Dasgupta and B

M. Dasgupta and B. K. El-Menoufi, JHEP12(2021), 158

2021
[20]

van Beekveld, M

M. van Beekveld, M. Dasgupta, B. K. El-Menoufi,et al.JHEP05(2024), 093

2024
[21]

Gustafson and U

G. Gustafson and U. Pettersson, Nucl. Phys. B306(1988), 746-758

1988
[22]

Andersson, G

B. Andersson, G. Gustafson, L. Lonnblad and U. Pettersson, Z. Phys. C43(1989), 625

1989
[23]

Frixione and B

S. Frixione and B. R. Webber, JHEP06(2002), 029

2002
[24]

Nason, JHEP11(2004), 040

P. Nason, JHEP11(2004), 040

2004
[25]

Hamilton, A

K. Hamilton, A. Karlberg,et al.JHEP03(2023), 224 [erratum: JHEP11(2023), 060]

2023
[26]

van Beekveld, S

M. van Beekveld, S. Ferrario Ravasio, J. Helliwell,et al.[arXiv:2504.05377 [hep-ph]]
[27]

Banfi, F

A. Banfi, F. A. Dreyer and P. F. Monni, JHEP03(2022), 135

2022
[28]

Becher, N

T. Becher, N. Schalch and X. Xu, Phys. Rev. Lett.132(2024) no.8, 8

2024
[29]

Heisteret al.[ALEPH], Eur

A. Heisteret al.[ALEPH], Eur. Phys. J. C35(2004), 457-486

2004

Pith tools

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

[1] [1]

van Beekveld, M

M. van Beekveld, M. Dasgupta,et al.Phys. Rev. Lett.134(2025) no.1, 011901

work page 2025

[2] [2]

C. Duhr, B. Mistlberger and G. Vita, Phys. Rev. Lett.129(2022) no.16, 162001

work page 2022

[3] [3]

X. Chen, T. Gehrmann,et al.Phys. Rev. Lett.128(2022) no.25, 252001

work page 2022

[4] [4]

Camarda, L

S. Camarda, L. Cieri and G. Ferrera, Phys. Lett. B845(2023), 138125

work page 2023

[5] [5]

M. Bahr, S. Gieseke, M. A. Gigg,et al.[Herwig], Eur. Phys. J. C58(2008), 639-707

work page 2008

[6] [6]

Bierlich, S

C. Bierlich, S. Chakraborty,et al.[Pythia], SciPost Phys. Codeb.2022(2022), 8

work page 2022

[7] [7]

Bothmannet al.[Sherpa], JHEP12(2024), 156

E. Bothmannet al.[Sherpa], JHEP12(2024), 156

work page 2024

[8] [8]

Dasgupta, F

M. Dasgupta, F. A. Dreyer,et al.JHEP09(2018), 033 [erratum: JHEP03(2020), 083]

work page 2018