Identifying regions for asymptotic expansions of amplitudes: fundamentals and recent advances

Yao Ma · 2025

Read Identifying regions for asymptotic expansions of amplitudes: fundamentals and recent advances by Yao Ma (2025) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2505-01368.txt.

Introduction

Yao Ma reviews how to find the regions that drive expansions of scattering amplitudes. His 2025 paper covers the method of regions, a workhorse for approximating Feynman integrals. The first step, listing all regions, decides whether the rest works. A geometry-based scheme using Newton polytopes handles many cases fast and well. Trouble grows with more loops or tricky graph shapes, where the list can hide missing entries. Ma sorts the hard cases, gives general strategies for them, and sums up what is known about region structures across expansion types.

Catalog record

CREATORYao Ma
YEAR2025
COLLECTIONPAPERS
FORMATTXT
FILE SIZE73 KB
VERSION1.0
LICENSEhttp://creativecommons.org/licenses/by/4.0/
RIGHTSCC BY
SOURCEarXiv

Source: https://arxiv.org/abs/2505-01368

Browse the full library · About the Tower of Babel

Full text

The opening of the work, from the library's hosted text:

Identifying regions for asymptotic expansions of amplitudes: fundamentals and recent advances Yao Ma Abstract This review paper discusses the identification of regions, a crucial first step in applying the “method-of-regions” technique. A systematic approach based on Newton polytope geometry has proven successful and efficient for many cases. However, obtaining the correct list of regions becomes increasingly subtle with higher loop numbers or specific Feynman graph topologies. This paper explores the scenarios where such subtleties arise, outlines general strategies to address them, and reviews the current understanding of region structures in various asymptotic expansions of Feynman integrals. keywords Feynman integral; asymptotic expansion; method of regions; infrared singularity; scattering amplitude PACS PACS numbers: 03.65. - w, 04.62.+v † † catchline: Vol. , No. () – 1 Introduction Evaluating multiloop Feynman integrals remains a central challenge in perturbative quantum field theory, particularly when multiple scales emerge in the external kinematics. A powerful approach to this problem is asymptotic expansion, which systematically disentangles the interplay of disparate scales. Consider a physical process characterized by a small scale \Lambda_{\textup{small}} (e.g. the mass of a light particle) and a large scale \Lambda_{\textup{large}} (e.g. the center-of-mass energy in a hard process). By introducing an expansion parameter \lambda\sim\Lambda_{\textup{small}}/\Lambda_{\textup{large}}\ll 1 , the integral \mathcal{I}(\lambda,\epsilon) admits a generalized series expansion of the form: \displaystyle\mathcal{I}(\lambda,\epsilon)\sim\sum_{\mu,k,n}c_{\mu,k,n}(\epsilon)\cdot\lambda^{\mu(\epsilon)}\cdot\textup{log}^{k}\lambda\cdot\lambda^{n} (1) where \epsilon denotes the dimensional regularization parameter, \mu(\epsilon) linearly depends on \epsilon , and the integers k,n track the logarithmic and power-law behaviors of \lambda , respectively. [ 1 ] Such expansions not only simplify calculations but also reveal the rich infrared structure of gauge theories by exposing hierarchies between hard, collinear, and soft dynamics that underlie factorization theorems and resummation techniques [ 2 , 3 , 4 , 5 , 6 , 7 , 8 , 9 ] . Among various techniques of doing asymptotic expansions, the one that we focus on in this paper is the “method of regions” (MoR) [ 10 ] , which is also referred to as the “expansion by regions” or “strategy of regions”. The basic statement is that, for the original integral \mathcal{I}(\lambda,\epsilon) above, there always exists a certain set of “regions” in the entire integration space, R_{1},R_{2},\dots,R_{n} , such that \mathcal{I} can be restored by summing over expansions around them. Namely, \displaystyle\mathcal{I}=\mathcal{I}^{(R_{1})}+\mathcal{I}^{(R_{2})}+\dots+\mathcal{I}^{(R_{n})}. (2) Note that for each term on the right-hand side, the integrand is modified through a corresponding Taylor expansion, while the integration measure remains as the entire integration space. As an important ingredient for the MoR, we note that evaluating each Feynman integral requires dimensional regularization, and all the scaleless integrals (e.g. \int d^{D}k(k^{2})^{\alpha} ) are set to zero. As an example, the expansion of the one-loop Sudakov form factor, as described in section 2.2 of Ref. [ 8 ] , treats one external momentum q to be the large energy scale, and p_{1} and p_{2} to be equally close to distinct lightcones: p_{1}^{2}\sim p_{2}^{2}\sim\lambda q^{2} with the expansion parameter \lambda\ll 1 . In this case, there are four regions in the loop-momentum space k^{\mu}=(k^{+},k^{-},\boldsymbol{k}_{\perp}) : • Hard region: k^{\mu}\sim q(1,1,1) ; • Collinear-1 region: k^{\mu}\sim q(1,\lambda,\lambda^{1/2}) ; • Collinear-2 region: k^{\mu}\sim q(\lambda,1,\lambda^{1/2}) ; • Soft region: k^{\mu}\sim q(\lambda,\lambda,\lambda) . By performing the corresponding expansions, the contribution from each region can be expressed in terms of Appell’s hypergeometric functions [ 11 ] . After summing up these contributions one reproduces the original Feynman integral to all orders . More examples of applying the MoR can be found in papers in the recent 20 years. Although the MoR has appeared as a powerful technique, such that it works for all the examples so far, two fundamental questions regarding its theoretical foundation remain to be answered: 1 . Is there a rigorous proof, or a counterexample, of the MoR? 2 . Suppose the MoR always works, how to identify the correct list of regions systematically? These two questions are, in fact, closely related. On the one hand, a rigorous answer to question 1 depends on 2 : any proof of the MoR must assume a correct choice of regions, while a counterexample of the MoR may simply stem from overlooking or double-counting certain regions. On the other hand, advances in addressing 1 can inform 2 . As Jantzen noted in Ref. [ 12 ] , the MoR is valid to all orders if the expansions satisfy specific conditions, such as domain compatibility, convergence, commutativity, etc. Throughout this paper, we shall focus on answering question 2 . Actually, an all-order identification of regions has been established for Feynman integrals with Euclidean kinematics [ 13 , 14 ] . A detailed demonstration can be found in Smirnov’s book [ 15 ] , where each region is characterized by a corresponding subgraph, termed the “asymptotically irreducible” subgraph. In this sense, expansions around regions can be seen as expansions regarding certain subgraphs. However, it remains highly nontrivial to extend this statement to Minkowski kinematics, where regions are usually determined based on heuristic examples or experience. This difficulty arises from two key challenges. First, the infrared structure in Minkowski space is more intricate: while in Euclidean spacetime k^{2}=0 requires all components of k^{\mu} to vanish, in Minkowski spacetime it can also occur when k^{\mu} lies on a lightcone, corresponding to jet-like configurations in infrared singularities. Second, the mode structure near singularities is complex. For the Sudakov form factor example above, one might wonder why additional modes (e.g., k^{\mu}\sim\lambda^{2}q ) are not involved in the MoR. A simple answer is that those scalings of k^{\mu} will lead to scaleless integrals which do not contribute to Eq. ( 2 ). However, for Feynman graphs at higher loop orders, a more refined and systematic analysis is often required. This paper reviews systematic approaches to identifying regions in the MoR, with a particular focus on recent advances from Refs. [ 16 , 17 , 18 ] . In Section 2 , we make some general remarks on the identification of regions. In particular, regions can be classified into two distinct types: facet regions and hidden regions . Section 3 specializes to the on-shell expansion for wide-angle scattering, presenting all-loop results for both region types. These findings are extended in Section 4 to some other wide-angle expansions, including the soft expansion , timelike-collinear expansion , and mass expansion for heavy-to-light decays. Section 5 examines the Regge-limit expansion for the 2\to 2 scattering process, highlighting the role of Glauber singularities in region analysis. We conclude with a summary and outlook in Section 6 . 2 General remarks on identifying regions This section establishes the foundation and introduces key concepts for subsequent analyses. In Section 2.1 , we review the geometric approach to regions based on Newton polytopes. With this knowledge, Section 2.2 then classifies regions of generic asymptotic expansions into two distinct types, facet regions and hidden regions, whose analyses employ different advanced strategies discussed in Section 2.3 . 2.1 Regions and Newton polytopes For any scalar Feynman integral corresponding to the graph G that is dimensionally regularized in D=4-2\epsilon dimensions, we consider the Lee-Pomeransky representation [ 19 ] : \mathcal{I}(\boldsymbol{s})=\mathcal{C}\cdot\int_{0}^{\infty}\left(\prod_{e\in G}\frac{dx_{e}}{x_{e}}\right)\cdot\left(\prod_{e\in G}x_{e}^{\nu_{e}}\right)\cdot\Big(\mathcal{P}(\boldsymbol{x};\boldsymbol{s})\Big)^{-D/2}, (3) where \mathcal{C} is an overall factor, e represents the edges of G , \boldsymbol{x} denotes the set of Lee-Pomeransky parameters, and \boldsymbol{s} denotes the

Read the complete text