On canonical differential equations for Calabi-Yau multi-scale Feynman integrals

Sara Maggio, Yoann Sohnle · 2025

Read On canonical differential equations for Calabi-Yau multi-scale Feynman integrals by Sara Maggio, Yoann Sohnle (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-2504-17757.txt.

Introduction

Sara Maggio and Yoann Sohnle derive canonical differential equations for Feynman integrals in two thousand twenty-five. The integrals have a Calabi-Yau geometry at their core and span many scales. A past method found canonical forms for simpler cases. The pair extend it to this harder class.

They first redo the sunrise integral with all masses distinct. They then give first time forms for two banana integrals with two distinct masses. The forms have a clean pole shape in the small parameter. Picard-Fuchs operators on the integrals ground the build. Appendices explain in steps why the extension works.

Catalog record

CREATORSara Maggio, Yoann Sohnle
YEAR2025
COLLECTIONPAPERS
FORMATTXT
FILE SIZE156 KB
VERSION1.0
LICENSEhttp://creativecommons.org/licenses/by/4.0/
RIGHTSCC BY
SOURCEarXiv

Source: https://arxiv.org/abs/2504-17757

Browse the full library · About the Tower of Babel

Full text

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

On canonical differential equations for Calabi-Yau multi-scale Feynman integrals Sara Maggio, Yoann Sohnle Abstract We generalise a method recently introduced in the literature, that derives canonical differential equations, to multi-scale Feynman integrals with an underlying Calabi-Yau geometry. We start by recomputing a canonical form for the sunrise integral with all unequal masses. Additionally, we compute for the first time a canonical form for the three-loop banana integral with two unequal masses and for a four-loop banana integral with two unequal masses. For the integrals we compute, we find an ϵ \epsilon -form whose connection has at most simple poles. We motivate our construction by studying the Picard-Fuchs operators acting on the integrals considered. In the appendices, we give a constructive explanation for why our generalisation works. Preprint: UUITP–14/25 BONN-TH-2025-16 1 Introduction Feynman integrals are central quantities in Quantum Field Theory, as they are essential building blocks for computing scattering amplitudes which are one of the bridges between theory and experiment Travaglini and others (2022) . Integrals of similar type have also appeared in the description of the scattering of compact objects in classical General Relativity in the so-called post-Minkowskian expansion Goldberger and Rothstein (2006) ; Porto (2016) ; Kälin and Porto (2020) ; Mogull et al. (2021) ; Kälin et al. (2020) ; Kälin et al. (2023) ; Jakobsen et al. (2022) ; Dlapa et al. (2024) . It is therefore necessary to develop sophisticated techniques which help us compute this type of integrals, in particular, by understanding their mathematical structure and underlying geometrical information. It turns out that at the core of all these integrals are classes of special functions related to complex manifolds where the simplest is the Riemann sphere where multiple polylogarithms (MPLs) Kummer (1840) ; Remiddi and Vermaseren (2000) ; Goncharov (1995) ; Goncharov (1998) ; Vollinga and Weinzierl (2005) ; Duhr et al. (2012) appear. However, diagrams needed for cutting-edge computations give rise to special functions beyond the ones that live on the Riemann sphere. These functions beyond MPLs are associated to periods of different geometries. The first example being the massive sunrise diagram in even dimensions which can be associated to an elliptic curve (which is isomorphic to a torus) Bloch and Vanhove (2015) ; Broadhurst (1990) ; Bauberger et al. (1995) ; Bauberger and Bohm (1995) ; Laporta and Remiddi (2005) . Recent progress has been made to go beyond the torus, on the one hand by studying Feynman integrals associated to higher genus Riemann surfaces Georgoudis and Zhang (2015) ; Duhr et al. (2025c) and their function space D’Hoker et al. (2023) ; Baune et al. (2024) ; D’Hoker et al. (2025) . On the other hand by studying Feynman integrals associated to higher dimensional Calabi-Yau manifolds Primo and Tancredi (2017) ; Klemm et al. (2020) ; Bourjaily et al. (2019) and their function space Bönisch et al. (2024) ; Duhr and Maggio (2025b) ; Duhr (2025) . Feynman integrals typically diverge and need to be regulated. The most commonly used regularisation scheme is dimensional regularisation ’t Hooft and Veltman (1972) ; Bollini and Giambiagi (1972) , where the integrals are computed in D=D_{0}-2\epsilon dimensions. They can then be expanded around \epsilon=0 . An extremely powerful technique to compute the integrals is through the differential equations (DEs) Kotikov (1991b) ; Remiddi (1997) ; Gehrmann and Remiddi (2000) , which are derived using integration-by-parts (IBPs) identities Tkachov (1981) ; Chetyrkin and Tkachov (1981) among Feynman integrals. This method is extremely convenient because we are not interested in the solution for generic values of \epsilon , but rather as a Laurent series in \epsilon . Moreover, in the last decade, it has been observed that it is often possible to find bases of Feynman integrals whose differential equations take a specific form, often referred to as canonical form Henn (2013) . The dependence on \epsilon can then be factorised out of the differential equation system, which also makes the analytic properties of their solutions close to \epsilon=0 completely manifest: each order in the Laurent series can be expressed as Chen iterated integrals Chen (1977) over the differential forms appearing in the connection. However, a completely general approach to finding canonical bases, even in the polylogarithmic case is not yet fully understood. Moreover, starting at two-loops, finding the canonical basis will involve a gauge transformation dependent on period functions of geometries beyond the Riemann sphere. In recent years, substantial effort has been dedicated to extend the definition of canonical differential equations to Feynman integrals beyond polylogarithms, and much progress has been made, in particular in the elliptic and one-parameter Calabi-Yau cases Görges et al. (2023) ; Pögel et al. (2022a) ; Adams et al. (2016) ; Adams et al. (2018b) ; Pögel et al. (2022b) ; Ahmed et al. (2024) ; Frellesvig (2022) ; Dlapa et al. (2023) ; Frellesvig and Weinzierl (2024) ; Chen et al. (2025) ; Pögel et al. (2023) . In particular, ref. Pögel et al. (2023) provides an ansatz to bring equal mass banana integrals, with underlying single-scale Calabi-Yau geometry, at any loop order, into \epsilon -factorised form. Moreover, very recently in ref. Duhr et al. (2025a) , a generalisation of the method of ref. Görges et al. (2023) has been developed for generic integrals with an underlying Calabi-Yau geometry (and beyond), providing explicit calculations for cutting edge Calabi-Yau one-scale integrals. The applicability of this algorithm has been explicitly demonstrated for many state-of-the-art problems containing integrals of elliptic type Görges et al. (2023) ; Becchetti et al. (2025a) ; Becchetti et al. (2025b) ; Duhr et al. (2024a) ; Forner et al. (2024) ; Marzucca et al. (2025) and for their generalisation to Calabi-Yau varieties Duhr et al. (2024a) ; Forner et al. (2024) ; Klemm et al. (2024) ; Driesse et al. (2024a) ; Driesse et al. (2024b) ; Duhr et al. (2025a) and higher-genus surfaces Duhr et al. (2025c) . In this paper, we propose a generalisation of the method introduced in ref. Pögel et al. (2023) , to multi-scale Calabi-Yau integrals. In the preparation of this work, we have realised that our generalisation coincides with the method from refs. Görges et al. (2023) ; Duhr et al. (2025a) , applied to multi-scale Calabi-Yau Feynman integrals. In fact, this is the first application of refs. Görges et al. (2023) ; Duhr et al. (2025a) to this class of integrals. To support our claim, we derive locally at the MUM-point, for the first time, a canonical differential equation for banana integrals with unequal masses. This paper is organised as follows. In section 2 we set our conventions and review the instruments that will be needed for the procedure we propose. Following this, in section 3 , we present our procedure for multi-variate Feynman integrals with underlying Calabi-Yau geometry. In section 4 , we recompute a canonical form for the unequal mass sunrise integral using the method presented before. Finally, in section 5 , we apply our construction to compute a canonical form for integrals with underlying multi-variate Calabi-Yau geometry, namely, the three-loop banana with two unequal masses (two-parameters K3) and the four-loop banana with two unequal masses (two-parameters CY three-fold). Additionally, we include three appendices. In the first two

Read the complete text