Local finiteness for real-virtual corrections to electroweak production in partonic collisions

We present a local subtraction scheme that enables the combined integration of loop momenta and the final-state parton phase space in real-virtual NNLO QCD corrections to cross sections for hadroproduction of electroweak and other colorless states. All initial- and final-state infrared singularities are subtracted at the integrand level in momentum space, yielding a locally finite integral ready for numerical integration in four dimensions. The subtraction terms are all based on the well-understood process of single-Higgs production. The core of our subtraction scheme relies on achieving local factorization in all infrared limits of real and virtual momenta. This necessitates systematic modifications of the original Feynman integrand for loop amplitudes, enabling gauge symmetry cancellations before performing integrations. Our approach provides an essential step toward NNLO cross-section calculations for hadron collider processes, where both loop and phase-space integrations are carried out numerically.

💡 Research Summary

The paper introduces a novel local subtraction framework that enables the simultaneous numerical integration of loop momenta and final‑state parton phase space for real‑virtual (RV) NNLO QCD corrections to the production of electroweak (EW) colour‑neutral states in hadronic collisions. Traditional NNLO calculations treat double‑real, double‑virtual and real‑virtual contributions separately, requiring intricate analytic manipulations to cancel infrared (IR) singularities across these pieces. This becomes prohibitive for processes involving many final‑state particles or multiple scales, such as multi‑boson production.

The authors’ key innovation is “local factorization”: they reorganize the integrand itself so that, in every soft or collinear limit of either the loop momentum k or the emitted gluon momentum p₃, the singular behaviour factorizes into universal splitting kernels already known from QCD factorization theorems. To achieve this, two technical steps are required.

1. Removal of loop‑polarization and power‑like singularities. In the q \bar q → EW + g channel, diagrams where a virtual gluon loop attaches to the quark‑gluon vertex generate contributions in which the real gluon’s polarization vector becomes proportional to the off‑shell loop momentum k rather than to p₃. These “loop‑polarization” terms produce logarithmic or even power divergences in the collinear limits (p₃‖p₁ or p₃‖p₂) and break factorization at the integrand level, although they cancel after integration. Similarly, self‑energy insertions on internal fermion lines generate spurious power‑like singularities when the internal fermion goes on‑shell while k remains off‑shell. The authors split the one‑loop amplitude into an “lp” part (containing the problematic terms) and a “nolp” part (already well‑behaved). They then modify the lp contribution by a specific momentum routing and by adding local counterterms that enforce longitudinal polarization of the emitted gluon and eliminate the power‑like behaviour. After this refinement, both parts are locally integrable and exhibit at most logarithmic collinear singularities.

2. Cross‑section‑based subtraction using single‑Higgs production as a template. The IR structure of any colour‑neutral EW final state is universal: it depends only on the identities of the incoming partons, not on the number or kinematics of the colour‑singlet bosons. Consequently, the authors can construct subtraction terms from the simplest process in the class—single‑Higgs production—where the real‑virtual integrand is already known and its IR limits are well studied. By scaling these template subtraction terms with the appropriate colour factors (leading‑color, sub‑leading‑color) and coupling constants, they obtain a set of local counterterms that cancel all initial‑state collinear poles (to be absorbed into PDFs) and all final‑state soft/collinear poles when the RV real and virtual pieces are summed.

The paper works out the full construction for two partonic channels: the quark‑antiquark (q \bar q) channel and the quark‑gluon (q g) channel. For each channel the authors:

• Enumerate all one‑loop diagram topologies (classes A–E) and assign a consistent momentum routing that respects the required factorization properties.
• Separate the amplitude into lp and nolp pieces, apply the refined loop‑polarization removal, and verify that the resulting integrand is locally finite in all IR limits.
• Decompose the colour structure into leading‑color (C_F T^a) and sub‑leading‑color (T^a C_A) contributions, and treat fermion‑loop (T^a ^2) pieces separately.
• Build explicit subtraction terms for each colour component, mirroring the structure of the single‑Higgs real‑virtual cross section.

Section 7 provides a detailed analytic proof of local factorization. The authors introduce a graphical notation to track soft and collinear limits of both the loop and real momenta, and they demonstrate that after the modifications the integrand reduces to a product of a universal eikonal factor (for soft gluons) or Altarelli‑Parisi splitting kernel (for collinear configurations) times a reduced Born‑like matrix element. They also discuss the “shift mismatch” problem—where certain non‑factorizable terms cancel only after shifting the loop integration variable—and show how their momentum routing eliminates the need for such shifts.

Section 8 presents numerical checks: the authors evaluate the integrand in randomly generated phase‑space points approaching each singular limit and confirm that the sum of the real‑virtual integrand and its local counterterms remains finite to machine precision.

The conclusions emphasize that the method yields a fully four‑dimensional, locally finite integrand for the RV contribution, allowing a straightforward Monte‑Carlo integration together with the double‑real and double‑virtual pieces. Because the subtraction terms are universal and derived from the simplest colour‑singlet process, the framework can be extended to more complex EW final states such as diphoton, Zγ, WW, and even multi‑boson production, as demonstrated in a companion study. The approach also dovetails with existing PDF factorization schemes, preserving the standard collinear subtraction structure while providing a new “cross‑section‑based” subtraction analogous to the DIS scheme for structure functions.

In summary, this work delivers a practical, analytically sound, and numerically implementable solution to the long‑standing problem of local IR cancellation in real‑virtual NNLO QCD corrections for electroweak colour‑neutral production. It paves the way for high‑precision predictions of a broad class of LHC processes, essential for stringent Standard Model tests and for uncovering subtle signals of new physics.