Joint LHC EFT Working Group & LHC Higgs Working Group Note: POPxf : An Exchange Format for Polynomial Observable Predictions Preprint: CERN-LHCEFTWG-2025-001 CERN-LPCC-2025-003 LHCHWG-2025-050 MITP-25-066 Affiliation: \naff ADipartimento di Fisica e Astronomia, Università di Bologna and INFN Bologna
Ilaria Brivio \aff A (ed.), Ken Mimasu \aff B (ed.), Peter Stangl \aff C (ed.), Anke Biekötter \aff KIT, Ana R. Cueto Gómez \aff UAM, Charlotte Knight \aff IC, Luca Mantani \aff IFIC, Eleonora Rossi \aff OX, Alejo N. Rossia \twoaff UNIPDINFNPD, Aleks Smolkovič \aff IJS · 2025
Read Joint LHC EFT Working Group & LHC Higgs Working Group Note: POPxf : An Exchange Format for Polynomial Observable Predictions Preprint: CERN-LHCEFTWG-2025-001 CERN-LPCC-2025-003 LHCHWG-2025-050 MITP-25-066 Affiliation: \naff ADipartimento di Fisica e Astronomia, Università di Bologna and INFN Bologna by Ilaria Brivio \aff A (ed.), Ken Mimasu \aff B (ed.), Peter Stangl \aff C (ed.), Anke Biekötter \aff KIT, Ana R. Cueto Gómez \aff UAM, Charlotte Knight \aff IC, Luca Mantani \aff IFIC, Eleonora Rossi \aff OX, Alejo N. Rossia \twoaff UNIPDINFNPD, Aleks Smolkovič \aff IJS (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-2511-17348.txt.
Introduction
Ilaria Brivio and nine co-authors introduce POPxf, a data format for publishing theory predictions in particle physics. Many predictions, especially in effective field theory, are polynomials in the model's parameters, yet they are often shared in ad hoc tables that hinder reuse. POPxf encodes such polynomial observable predictions in a structured, machine-readable form, recording all assumptions and metadata.
The format, developed on behalf of the LHC Higgs and Effective Field Theory working groups, handles errors and correlations flexibly enough to capture effects that depend on the parameters themselves. The authors aim at three practical gains: predictions others can reproduce, easier global fits that combine many measurements, and smoother reuse of results under new models. A public archive hosts the spec. The paper is infrastructure work, meant to standardize how theory reaches lab analyses.
Catalog record
| CREATOR | Ilaria Brivio \aff A (ed.), Ken Mimasu \aff B (ed.), Peter Stangl \aff C (ed.), Anke Biekötter \aff KIT, Ana R. Cueto Gómez \aff UAM, Charlotte Knight \aff IC, Luca Mantani \aff IFIC, Eleonora Rossi \aff OX, Alejo N. Rossia \twoaff UNIPDINFNPD, Aleks Smolkovič \aff IJS |
|---|---|
| YEAR | 2025 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 88 KB |
| VERSION | 1.0 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2511-17348
Browse the full library · About the Tower of Babel
Full text
The opening of the work, from the library's hosted text:
Joint LHC EFT Working Group & LHC Higgs Working Group Note: POPxf : An Exchange Format for Polynomial Observable Predictions Preprint: CERN-LHCEFTWG-2025-001 CERN-LPCC-2025-003 LHCHWG-2025-050 MITP-25-066 Affiliation: \naff ADipartimento di Fisica e Astronomia, Università di Bologna and INFN Bologna, Italy \naff BSchool of Physics and Astronomy, University of Southampton, Highfield, Southampton SO17 1BJ, United Kingdom \naff CInstitute of Physics, Johannes Gutenberg University Mainz, Staudingerweg 7, 55128 Mainz, Germany \naff KITKarlsruhe Institute of Technology, Institute for Theoretical Particle Physics, Wolfgang-Gaede-Straße 1, 76131 Karlsruhe, Germany \naff UAMUniversidad Autónoma de Madrid, Campus Universitario de Cantoblanco, Madrid, Spain \naff ICImperial College, London, United Kingdom \naff IFICInstituto de Fisica Corpuscular (IFIC), Universidad de Valencia-CSIC, E-46980 Valencia, Spain \naff OXUniversity of Oxford, Oxford, United Kingdom \naff UNIPDDipartimento di Fisica e Astronomia “G. Galilei”, Università di Padova, Via F. Marzolo 8, I-35131, Padova, Italy \naff INFNPDIstituto Nazionale di Fisica Nucleare, Sezione di Padova, Via F. Marzolo 8, I-35131, Padova, Italy \naff IJSJožef Stefan Institute, Jamova 39, 1000 Ljubljana, Slovenia Ilaria Brivio \aff A (ed.), Ken Mimasu \aff B (ed.), Peter Stangl \aff C (ed.), Anke Biekötter \aff KIT, Ana R. Cueto Gómez \aff UAM, Charlotte Knight \aff IC, Luca Mantani \aff IFIC, Eleonora Rossi \aff OX, Alejo N. Rossia \twoaff UNIPDINFNPD, Aleks Smolkovič \aff IJS Abstract We introduce the Polynomial Observable Prediction Exchange Format, POPxf , a structured, machine-readable data format for the publication and exchange of semi-analytical theoretical predictions in high energy physics. The format is designed to encode observables that can be expressed in terms of polynomials in model parameters, with particular emphasis on Effective Field Theory applications. All relevant assumptions and metadata are recorded explicitly, and the treatment of uncertainties and correlations is flexible enough to capture parameter-dependent effects. The format aims to improve reproducibility, facilitate global fits and reinterpretations, and streamline the use of theoretical predictions across the particle physics community. keywords Data format, Theoretical predictions, Effective Field Theory 0.1 Introduction Theoretical predictions play a central role in particle physics phenomenology, allowing the interpretation of experimental data in terms of the underlying model parameters. In many cases, such predictions can be expressed in terms of polynomials in these parameters. A prominent example is observables in an Effective Field Theory (EFT), where they depend on a set of Wilson coefficients. In particular, the amplitudes in an EFT are often linear in the Wilson coefficients, such that the associated observables, depending on squared amplitudes, are given in terms of quadratic polynomials in the Wilson coefficients. This structure arises generically in analyses within the Standard Model Effective Field Theory (SMEFT) and the Weak Effective Theory (WET) truncated at dimension six. Other examples beyond EFT include observables in flavour physics that depend on hadronic form factors, which enter amplitudes linearly and are themselves typically expressed using polynomial parameterisations. Despite the ubiquity of such polynomial expressions, they are rarely published in their entirety and often remain hidden in various public and private codebases. Even when predictions based on polynomial expressions are made accessible, this is often done using varying conventions, with insufficient metadata, and/or in formats that are not machine-readable. Such practices complicate reproducibility, hinder validation and cross-comparison, and lead to significant duplication of effort across the community. A standardised, machine-readable data format for sharing such predictions would address many of these issues. It would facilitate the reuse of theoretical predictions in global fits, reinterpretations, and future measurements. It would also help document and preserve the assumptions – such as EFT basis choices, renormalisation scales, input parameter values, or normalisation conventions – that are crucial for the correct use and interpretation of the results. This note proposes the Polynomial Observable Prediction Exchange Format, POPxf , a structured data format designed to encode polynomial parameterisations of observables as functions of model parameters, with a focus on EFT applications, while remaining general enough to support other use cases. The format supports arbitrary polynomial degree and is sufficiently general to describe both single-valued and multi-bin observables, entire sectors of related observables, as well as observables defined through functions of multiple polynomial components. Returning to the example of EFT predictions, where observables depend on a set of beyond-the-standard-model (BSM) Wilson coefficients, C_{i} , the scattering amplitude for a given process often takes the form \mathcal{A}=\mathcal{A}_{\mathrm{SM}}+\sum_{i}C_{i}\,\mathcal{A}_{i}, (1) so that the corresponding observable O , typically proportional to |\mathcal{A}|^{2} , becomes a quadratic polynomial in the C_{i} : O=O_{\mathrm{SM}}+\sum_{i}C_{i}\,O_{i}^{\mathrm{int}}+\sum_{i\leq j}C_{i}C_{j}\,O_{ij}^{\mathrm{quad}}. (2) Here, O_{\mathrm{SM}} is the Standard Model (SM) prediction, O_{i}^{\mathrm{int}} encodes the interference between SM and BSM amplitudes, and O_{ij}^{\mathrm{quad}} captures the pure BSM-squared contributions. This structure is general for dimension-six SMEFT and WET analyses, for which the expansion of the amplitude is truncated at \mathcal{O}(1/\Lambda^{2}) , where \Lambda is the cutoff parameter of the EFT. While quadratic polynomial structures are common, higher-order terms can arise in EFTs when including dimension-eight operators or computing beyond leading order. The proposed data format supports arbitrary polynomial degree to remain general and extensible. In general, model parameters such as Wilson coefficients can be complex. In such cases, observables may depend separately on the real and imaginary parts of the parameters. For example, a quadratic monomial C_{i}C_{j} must be interpreted as a combination of terms such as \mathrm{Re}(C_{i})\mathrm{Re}(C_{j}) , \mathrm{Re}(C_{i})\mathrm{Im}(C_{j}) , etc. The format explicitly supports such terms via a real-imaginary decomposition. Although we use the term ‘observable’ to refer to a quantity whose prediction is a single real number, in many cases the objects of interest are vector-valued quantities, i.e. sets of observables in the aforementioned sense. Examples include binned distributions and angular observables. In such cases, each bin or component is treated as a separate observable, and the predictions of the entire set of observables are encoded via vector-valued polynomial coefficients. The data format accommodates this naturally via array structures compatible with numerical data structures such as numpy or Mathematica arrays. Some observables are not themselves polynomials in the parameters but are functions of one or more polynomial expressions. Examples include ratios of decay widths or angular observables formed from linear combinations of angular coefficients. The data format supports such derived observables by allowing them to be defined in terms of functional combinations of underlying polynomials. Finally, the treatment of uncertainties and correlations in this format assumes that the observable is expressed as a polynomial in the model parameters. In many cases, the observable itself is already a polynomial – for example, cross sections or partial decay widths derived from squared amplitudes. In the case where the observable is defined as a function of one or more polynomials, it is expanded in the model parameters to yield a single polynomial expression, typically truncated at second order. This expansion defines the level at which parameter-dependent uncertainties and correlations are interpreted, and provides a consistent and efficient framework for uncertainty propagation [ Altmannshofer:2021qrr ] . The proposed data format is based on the JSON standard, a lightweight, text-based data format that is both human-readable and natively supported in most programming languages and computer algebra systems, particularly in python and Mathematica , which are used extensively in the particle physics community. It provides a simple and transparent way to encode central values, uncertainties and correlations as well as all relevant metadata in a structured, hierarchical format. Specific design goals include: