Infrared singularities and the collinear limits of multi-leg scattering amplitudes

Sebastian Jaskiewicz

REVIEW 4 major objections 5 minor 32 references

Infrared singularities and the collinear limits of multi-leg scattering amplitudes

T0 review · 4 major / 5 minor · reviewed 2026-08-07 · deepseek-v4-flash

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

Pith's one-line read Two-particle collinear limits alone guarantee multi-particle collinear factorisation through four loops.

desk verdict A clear proceedings summary of the author's own parent paper; the results are real but deferred to [1], so this text alone can't support them. read the letter →

arxiv 2608.06157 v1 pith:TQDQ6SVM submitted 2026-08-06 hep-ph

Sebastian Jaskiewicz This is my paper · ORCID

classification hep-ph

keywords softanomalousdimensioncollinearfactorisationmulti-legamplitudesinfraredsingularitiesfour-loopcorrectionssplittingmassivecolouredpartonsconformalcross-ratios

verification ladder T0 review T1 audit T2 compute T3 formal

The pith

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

The reading

This paper asks whether the factorised structure of scattering amplitudes in two-particle collinear limits is enough to guarantee factorisation when several particles become collinear simultaneously. By inserting the complete four-loop soft anomalous dimension — the operator that controls infrared poles in QCD amplitudes — into the difference that defines the splitting-amplitude anomalous dimension, the author shows that once all two-particle collinear limits are factorised, every multi-particle collinear limit of massless amplitudes is automatically factorised. No new constraints on the soft anomalous dimension are gained from multi-collinear limits of massless partons through four loops. If one external particle is massive, however, the three-particle collinear limit does impose a new condition, given in eq. (7), that relates the massive function $F^{h3}$ to the massless four-loop function $F$.

What carries the argument

The central object is the soft anomalous dimension $\Gamma_n$, which organises the infrared singularities of the factorised amplitude. The operative identity is eq. (6): the soft anomalous dimension of the splitting amplitude is $\Gamma_{Sp,m} = \Gamma_n - \Gamma_{n-m+1}$ with the parent colour charge $T_P$ replaced by the sum of the collinear charges; strict collinear factorisation requires this difference to depend only on the collinear particles. The paper inserts the four-loop structure of $\Gamma_n$, whose non-dipole terms are functions of conformally invariant cross-ratios $\beta_{ijkl}$ (and massive variants $r_{ijI}$), and tracks the cancellation of the reference-parton dependence. The new massive constraint eq. (7) follows from taking three massless momenta $p_a, p_b, p_c$ collinear in the massive function $F^{h3}$.

What would settle it

Compute a four-loop three-particle splitting-amplitude soft anomalous dimension directly from Feynman diagrams and compare it with the difference $\Gamma_n - \Gamma_{n-m+1}$ used here; any mismatch would disprove the claim. For the massive case, an independent four-loop calculation of the collinear limit of $F^{h3}$ that disagrees with eq. (7) would refute the new constraint.

Watch

Extended reading notes

Core claim

The central claim is that through four loops, strict collinear factorisation in all two-particle collinear limits is sufficient to guarantee strict collinear factorisation in any multi-particle collinear limit of a massless amplitude. The argument goes by substituting the known four-loop structure of the soft anomalous dimension $\Gamma_n$ into the defining relation $\Gamma_{Sp,m} = \Gamma_n - \Gamma_{n-m+1}$ with the parent colour charge replaced by the sum of the collinear charges, and showing that all dependence on the non-collinear reference partons cancels. For massless partons the constraint is independent of the total number of legs and of the number $m$ of collinear particles: all information is already contained in the two-particle limits. When one massive coloured parton is present, the same construction, applied to three massless collinear particles, yields a new constraint, eq. (7), which fixes the collinear behaviour of the massive three-particle function $F^{h3}$ and agrees with the previously computed small-mass limit.

Load-bearing premise

The argument assumes that eq. (3) is the complete four-loop soft anomalous dimension, including every colour structure and collinear-singular term; if a term is missing, the cancellations that prove multi-particle factorisation could fail.

Editorial extensions

If this is right

Massless bootstrap programs need only enforce two-particle collinear factorisation up to four loops; multi-collinear limits then follow automatically.
The two-particle constraint is universal in the number of external legs: checking $\Gamma_n - \Gamma_{n-1}$ for any $n$ provides no additional information beyond the fixed-$n$ checks used previously.
For amplitudes with one massive coloured parton, the three-particle collinear limit gives a new, independent condition, eq. (7), on the massive function $F^{h3}$.
The new condition is consistent with the known small-mass limit of $F^{h3}$, so the massive collinear limit cross-checks existing four-loop results.
Strict collinear factorisation holds for massless multi-collinear limits at three and four loops, confirming that the cancellation mechanism in eq. (6) is realised by the known colour and kinematic structure.

Reading between the lines

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

A natural extrapolation is that for massless amplitudes the pattern continues beyond four loops: any violation of multi-collinear factorisation would require a new colour structure absent from eq. (3), so a five-loop check of that equation is a direct probe of the pattern's stability.
Eq. (7) can be read as a bootstrap condition rather than only a check: any proposed four-loop massive contribution that does not reduce to the required massless function in the collinear limit would be excluded by collinear factorisation.
The same difference-of-anomalous-dimensions machinery could be applied to spacelike collinear limits, where factorisation breaking and coherence-violating logarithms are known to appear; the paper lists this as an outlook rather than a result.

Share X Bluesky LinkedIn Reddit HN

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

Dear colleague,

Quick take: this is a proceedings write-up of the author's own parent paper (Duhr, Gardi, Jaskiewicz et al., JHEP 02 (2026) 173). It states two results: (1) through four loops, satisfying all two-particle collinear constraints on the massless soft anomalous dimension guarantees factorization in every multi-particle collinear limit, so no new bootstrap information comes from the multi-particle limit; (2) with a massive colored parton, the three-particle limit gives a new constraint, eq. (7), relating F^{h3} to the massless quadruple-correlation function F. The paper is honest about its own status: derivations are explicitly deferred to [1], and the four-loop structure in eq. (3) is given only schematically.

What's good: the logic is laid out clearly. Eq. (6) is the right tool, and the paper explains why strict collinear factorization forces the right-hand side to be independent of spectator partons. The massive case is genuinely interesting—eq. (7) is a concrete, compact constraint that could pin down unknown four-loop functions. The reference list is appropriate, and the heavy self-citation is transparent rather than hidden.

Soft spots: the central claim is not backed by anything in this text. Completeness of eq. (3)—the assumption that those seven terms exhaust the four-loop soft anomalous dimension—is not established here; the paper refers to [6] and [7] but doesn't show the forms. The stress-test worry is valid: if there is a four-loop term that vanishes in every two-particle limit but not in a multi-particle limit, the massless conclusion would break. Nothing in this proceedings rules that out. Same for eq. (7): it's stated as a computed limit, but the calculation is elsewhere. So as a self-contained paper, it's thin. As a proceedings, it's normal, and the reader is told where to look.

My judgment: the physics is probably right if the parent paper survives scrutiny, but this text alone isn't enough to assess it. I would not send this to a referee as a standalone article; it's a conference contribution pointing to the real paper. If the venue expects original content, desk reject. If it's a genuine proceedings, light editorial check is fine.

Recommendation: don't treat this as an independent result; cite the parent publication. For a reading group, it could be a 20-minute overview, but only as an entry point to [1].

Referee Report

4 major / 5 minor

Summary. This proceedings contribution, based on the author's parent publication [1], studies timelike multi-particle collinear limits of n-parton scattering amplitudes. It claims that through four loops, all two-particle collinear-factorization constraints on the massless soft anomalous dimension imply factorization in any multi-particle collinear limit, so multi-collinear limits provide no additional bootstrap constraints for massless amplitudes; and that in the presence of one massive coloured parton, the three-massless-particle collinear limit yields a new constraint on the function F^{h3}, displayed in Eq. (7). The paper presents the soft anomalous dimension decomposition (Eq. (3)) and the splitting soft-anomaly relation (Eq. (6)), then states its conclusions with the derivations deferred to the parent publication [1].

Significance. If the results are correct, they are useful for the bootstrap program for the soft anomalous dimension: they identify which collinear limits can constrain unknown four-loop colour structures and which cannot. The manuscript is clearly written and transparent about its status as a proceedings summary; it correctly situates the work in the existing literature and does not pretend to contain full derivations. The significance is nevertheless conditional: because all computational evidence is deferred to [1], the present text alone cannot establish the central claims, and the conclusions are only as strong as the completeness of Eq. (3) and the correctness of the calculations in [1].

major comments (4)

[Sec. 3, after Eq. (6)] The central massless claim is not supported within the manuscript: the text states that the calculations are 'lengthy and are not presented here' and refers to [1], and Eq. (3) is given only schematically ('The explicit forms ... can be found in ... [1]'). As a result, neither the completeness of the four-loop decomposition nor the cancellation that removes spectator dependence from Gamma_Sp,m can be checked from the present text. Please include the explicit forms of the relevant contributions to Gamma_n, or add an appendix with a proof of the multi-collinear factorization statement, or at minimum give a detailed outline of the calculation with the key intermediate steps.
[Sec. 3, Eq. (7)] The new massive-parton constraint is stated without derivation. The middle expression in Eq. (7) uses arguments (beta_abIc, beta_acIb; r_bcI) that are not defined anywhere in the text, and the limiting procedure r_bcI -> 0 is not described. The equality to 4F(beta_ablc, beta_aclb) in the triple-collinear limit is asserted rather than demonstrated. Please supply the derivation, define all variables, and clarify how this constraint follows from strict collinear factorization.
[Sec. 3, first bullet] The claim of universality of the two-particle collinear constraint -- namely, that considering Gamma_n - Gamma_{n-1} for arbitrary n gives no new information beyond the small-n cases considered previously -- is stated without proof. This universality is load-bearing for the conclusion that multi-collinear limits add no new constraints for massless amplitudes. Please include a proof sketch or an explicit statement of where in [1] this is established.
[Sec. 3, Eq. (3)] The argument depends on Eq. (3) being a complete list of contributions to the massless soft anomalous dimension through four loops. The text says only that the four-loop structure was 'investigated' in [6] and writes the terms only schematically. If there exists a four-loop contribution absent from Eq. (3) that vanishes in all two-particle collinear limits but develops a nonvanishing limit when three or more partons are simultaneously collinear, the central claim would fail. Please state the basis on which Eq. (3) is taken to be exhaustive, and explain why no such term can occur.

minor comments (5)

[Abstract and Sec. 1] The text contains spacing/OCR artifacts such as 'n-particle', 'n-massless', and 'the ϵ→0 limit'; these should be typeset correctly.
[Eq. (3)] The labels 4T-3L, 4T-4L, Q4T-2,3L, 5T-4L, and 5T-5L are never defined. Please spell out the meaning of these colour/loop designations at first use.
[Eq. (7)] The notation for the arguments of F^{h3} is inconsistent: the limit is written with r_abI, r_acI, r_bcI, but the next expression uses beta_abIc, beta_acIb and a semicolon before r_bcI. Please unify the notation and define every argument.
[Sec. 3, Eqs. (5)-(6)] The splitting-amplitude notation is inconsistent: 'Sp m' appears in Eq. (5), while 'Sp,m' and 'Gamma_Sp,m' appear in the surrounding text and Eq. (6). Please use a single convention.
[References] Because this is a proceedings summary, it would help the reader if the text indicated which specific sections or equations of [1] contain the derivations of the claims made in Sec. 3.

Circularity Check

1 steps flagged · score 5.0 of 10

Headline claims are imported wholesale from the author's own parent publication [1]; they are genuine consistency checks (not tautologies), so the circularity is load-bearing self-citation rather than definitional.

self citation load bearing [Sec. 3, paragraph after eq. (6); Sec. 2, paragraph after eq. (3)]
"The explicit forms of these terms are rather lengthy, but can be found in a suitable form for this discussion in Sec. 2 of the publication on which this proceeding is based 1. ... The calculations are lengthy and are not presented here, however detailed derivation can be found in the parent publication 1. ... Here, we state the main conclusions of this work 1:"

Every headline claim of Sec. 3 — the universality of the two-particle constraint, the massless result that multi-collinear limits 'are directly satisfied at three and four loops, as soon as all two-particle collinear limits are satisfied', and the new massive constraint eq. (7) — is stated, not derived, in this text. The input terms of eq. (3) live only in 'Sec. 2 of the publication on which this proceeding is based [1]', and the 'detailed derivation can be found in the parent publication [1]'. Reference [1] is the author's own paper (the acknowledgments thank exactly the [1] collaborators), so the central argument terminates in a self-citation: the conclusion holds iff the unshown parent-paper calculation is correct; it cannot be checked here. Not a definitional tautology — eq.

full rationale

No self-definitional step, no fitted input renamed as a prediction, and no uniqueness theorem imported from the authors. The soft anomalous dimension in eq. (3) is taken from external derivations ([5], [6], [7], [8]); the paper itself flags a completeness limitation by saying the four-loop structure was only 'investigated' [6] and that eq. (3) is 'schematic', so the four-loop conclusion inherits an unproven-completeness assumption — that is a correctness risk, not circularity. The massive F^{h3} comes from [11], and the new massive constraint eq. (7) is externally corroborated by the small-mass-limit computation [10] (Liu & Schalch), so the constraint is a genuine consistency condition rather than a tautology. The circularity that does exist is pattern self_citation_load_bearing: the entire derivation of the central claims (universal two-particle constraint; massless multi-collinear factorisation; eq. (7)) is deferred to the author's own parent publication [1], with the text saying the calculations 'are not presented here' and the 'detailed derivation can be found in the parent publication [1]'. From the text alone the central result cannot be verified; the claim is exactly as strong as the unshown [1] calculation. Score 5: the central argument's verification reduces to a self-citation (above 4), but the claims retain independent content and external corroboration, so they are not forced by definition (below 6).

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

The paper introduces no free parameters or new entities. Its central claims rest on the imported structure of the soft anomalous dimension and on the strict collinear factorisation criterion, both taken from the prior literature and the parent publication.

assumptions (4)

domain assumption The soft anomalous dimension has the structure of eq. (3) through four loops, with contributions from dipole formula, 4T, Q4T, 5T terms and massive corrections.
Used throughout Sec. 3 to evaluate Γ_{Sp,m}. Imported from refs [5,6,7,8,10,11].
domain assumption The factorization in eq. (1) and the RG solution in eq. (2) are valid.
Standard in the IR singularity literature; the paper relies on it without proof.
domain assumption Strict collinear factorisation, expressed by eq. (6), is the correct physical criterion; i.e., Γ_{Sp,m} must be independent of all non-collinear partons.
This is the constraint that drives the computation; the paper assumes this is what must hold.
domain assumption The limits of the CICRs and massive variables are taken in the standard way, and analytic continuation into the collinear limit is well-defined.
Needed for eqs. (4) and (7).

how reviews work

0 comments

Cite this review

Pith. "Pith review of Infrared singularities and the collinear limits of multi-leg scattering amplitudes." pith.science (2026). https://pith.science/paper/TQDQ6SVM

@misc{pith2026260806157,
  author       = {Pith},
  title        = {Pith review of: Infrared singularities and the collinear limits of multi-leg scattering amplitudes},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/TQDQ6SVM}},
  note         = {Machine review of arXiv:2608.06157}
}

abstract

Scattering amplitudes admit a factorised structure in special kinematic limits, such as the soft and collinear limits. In this work, we investigate the multi-particle collinear limits of massless amplitudes at high perturbative orders, focusing on the exploration of the mechanisms via which strict collinear factorisation of $n$-particle scattering amplitudes is realised when $m$ particles become collinear. We show through four loops that the requirements on the structure of the massless soft anomalous dimension that are imposed by strict collinear factorisation in all two-particle collinear limits are enough to guarantee factorisation also in any multi-particle collinear limit. Demanding that strict collinear factorisation of massless partons is satisfied also for amplitudes that contain a massive coloured particle, we derive new constraints on the soft anomalous dimension by considering the collinear limit of three massless particles.

Figures

Figures reproduced from arXiv: 2608.06157 by the authors.

**Figure 2.** Diagram showing the multi-particle timelike collinear limit of an [PITH_FULL_IMAGE:figures/full_fig_p003_2.png] view at source ↗

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

32 extracted references · 29 canonical work pages

[1]

C. Duhr, E. Gardi, S. Jaskiewicz,et al., JHEP02(2026), 173

work page 2026
[6]

Becher, M

T. Becher, M. Neubert, JHEP01(2020), 025

work page 2020
[7]

J. Henn, G. Korchemsky, B. Mistlberger, JHEP04(2020), 018

work page 2020
[2]

Becher, M

T. Becher, M. Neubert, JHEP06(2009), 081

work page 2009
[3]

Gardi, L

E. Gardi, L. Magnea, Nuovo Cim. C32N5-6(2009), 137–157

work page 2009
[4]

Agarwal, L

N. Agarwal, L. Magnea, C. Signorile-Signorile, A. Tripathi, Phys. Rept.994(2023)

work page 2023
[5]

Almelid, C

Ø. Almelid, C. Duhr, E. Gardi, Phys. Rev. Lett.117(2016), 172002

work page 2016
[8]

Falcioni, E

G. Falcioni, E. Gardi, N. Maher, C. Milloy, L. Vernazza, JHEP03(2022), 053

work page 2022

Show all 32 references

[9]

Gardi, L

E. Gardi, L. Magnea, JHEP03(2009), 079

2009
[10]

Z. L. Liu, N. Schalch, Phys. Rev. Lett.129(2022), 232001

2022
[11]

Gardi, Z

E. Gardi, Z. Zhu, [arXiv:2510.27567]
[12]

Berends, W

F. Berends, W. Giele, Nucl. Phys. B313(1989), 595–633

1989
[13]

Mangano, S

M. Mangano, S. Parke, Phys. Rept.200(1991), 301–367

1991
[14]

Z. Bern, G. Chalmers, Nucl. Phys. B447(1995), 465–518

1995
[15]

Kosower, Nucl

D. Kosower, Nucl. Phys. B552(1999), 319–336

1999
[16]

Dixon, E

L. Dixon, E. Gardi, L. Magnea, JHEP02(2010), 081

2010
[17]

Catani, D

S. Catani, D. de Florian, G. Rodrigo, JHEP07(2012), 026

2012
[18]

Almelid, C

Ø. Almelid, C. Duhr, E. Gardi, A. McLeod, C. White, JHEP09(2017), 073

2017
[19]

Forshaw, M

J. Forshaw, M. Seymour, A. Siodmok, JHEP11(2012), 066

2012
[20]

Schwartz, K

M. Schwartz, K. Yan, H. X. Zhu, Phys. Rev. D96(2017), 056005

2017
[21]

Cieri, P

L. Cieri, P. Dhani, G. Rodrigo, Phys. Rev. D113(2026), L031506

2026
[22]

C. Duhr, A. Venkata, C. Zhang, Phys. Rev. Lett.135(2025), 241601

2025
[23]

J. Henn, R. Ma, Y. Xu, K. Yan,et al., Phys. Rev. D112(2025), 076003

2025
[24]

Buccioni, H

F. Buccioni, H. Fang, K. Yan, [arXiv:2603.27123]
[25]

Becher, P

T. Becher, P. Hager, S. Jaskiewicz,et al., Phys. Rev. Lett.134(2025), 061901

2025
[26]

Becher, P

T. Becher, P. Hager, S. Jaskiewicz,et al., JHEP01(2026), 024

2026
[27]

Nabeebaccus, J

S. Nabeebaccus, J. Schoenleber,et al., Phys. Rev. D111(2025), 034040

2025
[28]

Banfi, J

A. Banfi, J. Forshaw, J. Holguin, Phys. Rev. Lett.136(2026), 221901

2026
[29]

Dasgupta, A

M. Dasgupta, A. Fraley, P. Monni, S. Nabeebaccus, [arXiv:2511.14681]
[30]

Becher, P

T. Becher, P. Hager, M. Neubert, D. Schwienbacher, [arXiv:2603.12383]
[31]

W. Chen, E. Gardi, R. Ma, Y. Ma, Y. Zhang, Z. Zhu, [arXiv:2607.15126]

arXiv
[32]

Barcaro, A

D. Barcaro, A. Gao, J. Gaunt, A. Pathak, [arXiv:2607.14259]

arXiv

Pith tools

Reviewed August 7, 2026 · model on record in the stance chip above.

[1] [1]

C. Duhr, E. Gardi, S. Jaskiewicz,et al., JHEP02(2026), 173

work page 2026

[2] [6]

Becher, M

T. Becher, M. Neubert, JHEP01(2020), 025

work page 2020

[3] [7]

J. Henn, G. Korchemsky, B. Mistlberger, JHEP04(2020), 018

work page 2020

[4] [2]

Becher, M

T. Becher, M. Neubert, JHEP06(2009), 081

work page 2009

[5] [3]

Gardi, L

E. Gardi, L. Magnea, Nuovo Cim. C32N5-6(2009), 137–157

work page 2009

[6] [4]

Agarwal, L

N. Agarwal, L. Magnea, C. Signorile-Signorile, A. Tripathi, Phys. Rept.994(2023)

work page 2023

[7] [5]

Almelid, C

Ø. Almelid, C. Duhr, E. Gardi, Phys. Rev. Lett.117(2016), 172002

work page 2016

[8] [8]

Falcioni, E

G. Falcioni, E. Gardi, N. Maher, C. Milloy, L. Vernazza, JHEP03(2022), 053

work page 2022