Source-linked AI summary
Revisiting parton evolution and the large-x limit
Yu. L. Dokshitzer, G. Marchesini, G. P. Salam
TL;DR
The paper addresses how to explain a suggestive relation among large-N three-loop anomalous dimensions and proposes a reformulation of parton evolution based on nonlinear combinations of lower loops. This model explains the space-like relation and agrees with the available two-loop time-like result, while further higher-order structure remains unresolved.
Problem
The paper asks whether the observed large-N relation among three-loop anomalous-dimension coefficients has a structural explanation, amid the broader challenge of simplifying their higher-loop structure.
Source: §1 · pp. 1–2Method
It reformulates the evolution equation so that higher-loop anomalous dimensions can partly inherit structure from nonlinear combinations of lower loops.
Source: §1 · pp. 1–2Results
The model explains the space-like MVV observation and agrees with the available two-loop time-like result.
Source: §2 · p. 4Takeaways & Limitations
The proposed splitting functions would be more compact, vanish at large x beyond first loop, and be identical for space- and time-like evolution.
Source: §1 · p. 2The model's space-like three-loop D3 result mismatches the full result, and additional three-loop structures remain unexplained, including terms whose origin may depend on understanding off-diagonal splittings.
Source: §3 · p. 5
Abstract
from arXiv · showhide
This remark is part of an ongoing project to simplify the structure of the multi-loop anomalous dimensions for parton distributions and fragmentation functions. It answers the call for a "structural explanation" of a "very suggestive" relation found by Moch, Vermaseren and Vogt in the context of the x->1 behaviour of three-loop DIS anomalous dimensions. It also highlights further structure that remains to be fully explained.
1 Introduction
The paper proposes that nonlinear combinations of lower-loop anomalous dimensions may explain some higher-loop structure, motivating a reformulation of parton evolution. The proposed splitting functions would be more compact, vanish at large x beyond first loop, and match between space- and time-like evolution.
1 Introduction
- The distributions evolve through an anomalous-dimension matrix, whose perturbative calculations become increasingly cumbersome beyond leading order.Splitting functions are conventionally defined as inverse Mellin transforms of the corresponding anomalous dimensions.Source: §1 · p. 1
- The authors suspect higher-loop anomalous-dimension structure can partly be understood as inherited from nonlinear combinations of lower loops.They frame this as a possible reformulation of the evolution equation, rather than an established explanation.Source: §1 · pp. 1–2
- The proposed reformulation would make splitting functions more compact and have them vanish at large x beyond first loop while matching between space- and time-like evolution.The latter property would restore Gribov-Lipatov reciprocity, broken beyond first loop in the standard formulation.Source: §1 · p. 2
- In the physical-coupling formulation, the two-loop splitting function vanishes as (1−x) when x approaches 1, consistent with a picture assigning higher-loop terms to quantum multiparton fluctuations.The cited soft-radiation argument assigns soft singularities to the first loop.Source: §1 · p. 2
- The authors remain far from understanding the full multi-loop structure, notably because of complications from off-diagonal transitions.Their investigation focuses on implications of the proposed relation for the large-N structure of three-loop anomalous dimensions.Source: §1 · p. 2
- Moch, Vermaseren and Vogt observed related three-loop coefficients in non-singlet quark and diagonal singlet quark-quark and gluon-gluon splitting functions.The reported large-N relations include C_a^3 = 2A_a^1A_a^2 and C_a^2 = (A_a^1)^2.Source: §1 · pp. 2–3
2 MVV relation
The paper tests whether a reformulated evolution equation can account for the large-N coefficient relation observed by Moch, Vermaseren and Vogt. Its model explains the space-like relation and agrees with the available two-loop time-like result.
2 MVV relation
- The model approximates the splitting function by the physical coupling times the one-loop splitting function to study the x→1 limit.It rewrites the equation to handle the correlated z and Q² dependences before taking its Mellin transform.Source: §2 · p. 3
- The coefficient relation is derived from an approximation intended specifically for the x→1 limit.The derivation represents the resulting relation in terms of coefficients C_n.Source: §2 · pp. 3–4
- The Mellin-space model uses γ1(N), the transform of the first-order splitting function, and applies ∂t only to D_N, not to α.The authors use its iterative solution to derive an all-order model for the anomalous dimension.Source: §2 · p. 3
- To recover the MVV relation, the derivation keeps γ1 = −A1 ln N + O(1) and retains the term proportional to its N-derivative times γ1.The authors identify A1 with the corresponding A_a coefficient before writing the all-order relation between C_a and A_a.Source: §2 · pp. 3–4
- For space-like evolution, the reformulated model explains the MVV observation; for time-like evolution, its two-loop result agrees with the available calculation.The reported relation is expressed in terms of expansion coefficients C_n.Source: §2 · p. 4
3 Pushing our luck
The large-N test finds agreement for D2 and some logarithmically enhanced terms, but a space-like D3 mismatch and further unresolved structures remain. The analysis extends the endpoint expansion while noting where its expected control may fail.
3 Pushing our luck
- The authors expect control over the 1/N term, but anticipate that the expansion will fail at 1/N² because it corresponds to quantum terms vanishing as 1−x.The expected boundary may be modified by logarithms.Source: §3 · p. 4
- The x-space expansion uses x/(1−x), as motivated by Low’s theorem [9], and retains γ1’s constant term because the nonlinear construction multiplies it by γ̇1∼1/N when calculating D.The all-order extension for B follows by substituting αγ1 with α(γ1−B1)+B; B disappears except in undotted γi factors.Source: §3 · p. 4
- D2 agrees across space-like and time-like evolution and quark and gluon channels, but the predicted space-like D3 differs from the full result for both quarks and gluons.The authors note that replacing the physical-coupling derivative with an MS-coupling derivative would remove the mismatch, but give no reason to prefer that substitution.Source: §3 · pp. 4–5
- Agreement holds for non-singlet α²s(1−x)ln²(1−x) terms, while diagonal-singlet agreement at α³s(1−x)ln²(1−x) is limited to nf-independent parts.The diagonal singlet also has an unexpected nf-proportional α³s(1−x)ln³(1−x) contribution.Source: §3 · p. 5
- Structures in the α³s(1−x)ln(1−x) terms remain unexplained, and the origin of the nf-proportional singlet ln³(1−x) contribution may require understanding higher-order off-diagonal splittings.The authors propose soft-gluon universality and space/time-like reciprocity as possible sources of simplicity, while noting that the picture’s consistency remains unresolved.Source: §3 · p. 5
Appendix
The appendix leaves the proposed relation’s extension to non-diagonal transitions unresolved and reports that two-loop gluon–gluon universality holds for only two color structures.
Appendix
- The authors cannot generalize relation (3) to non-diagonal transitions, whose large-x divergences scale as α_s^n ln^(2n−2)(1−x).They suspect these transitions pick up residues from ratios of non-cancelling divergent Sudakov exponents and singular integrals of those ratios.Source: Appendix · pp. 5–6
- In the supersymmetric case, identical quark and gluon Sudakov exponents cancel, consistent with most logarithmic enhancements cancelling.The proposed residues from non-cancelling Sudakov exponents are therefore consistent with the reported supersymmetric cancellation.Source: Appendix · p. 6
- At two loops, the gluon–gluon splitting function is universal for only the C_A^2 and C_A n_f color structures.For the remaining C_F n_f structure, the analogue of relation (4) links gluon–gluon and singlet quark–quark splittings.Source: Appendix · p. 6