Publications

Order by: publication date number of citations

• 2B or not 2B, a study of bottom-quark-philic semi-visible jets

  Deepak Kar, Wandile Nzuza, Sukanya Sinha

  SciPost Phys. Core 7, 071 (2024) · published 13 November 2024 | Toggle abstract · pdf

  Semi-visible jets arise in strongly interacting dark sectors, resulting in jets overlapping with the missing transverse momentum direction. The implementation of semi-visible jets is done using the Pythia Hidden Valley module to mimic the QCD sector showering in so-called dark shower. In this work, only heavy flavour Standard Model quarks are considered in dark shower, resulting in a much less ambiguous collider signature of semi-visible jets compared to the democratic production of all five quark flavours in dark shower. The constraints from available searches on this signature are presented, and it is shown the signal reconstruction can be improved by using variable-radius jets. Finally a search strategy is suggested.

• Precision-machine learning for the matrix element method

  Theo Heimel, Nathan Huetsch, Ramon Winterhalder, Tilman Plehn, Anja Butter

  SciPost Phys. 17, 129 (2024) · published 8 November 2024 | Toggle abstract · pdf

  The matrix element method is the LHC inference method of choice for limited statistics. We present a dedicated machine learning framework, based on efficient phase-space integration, a learned acceptance and transfer function. It is based on a choice of INN and diffusion networks, and a transformer to solve jet combinatorics. We showcase this setup for the CP-phase of the top Yukawa coupling in associated Higgs and single-top production.

  6 citations

• Les Houches guide to reusable ML models in LHC analyses

  Jack Y. Araz, Andy Buckley, Gregor Kasieczka, Jan Kieseler, Sabine Kraml, Anders Kvellestad, Andre Lessa, Tomasz Procter, Are Raklev, Humberto Reyes-Gonzalez, Krzysztof Rolbiecki, Sezen Sekmen, Gokhan Unel

  SciPost Phys. Comm. Rep. 3 (2024) · published 6 November 2024 | Toggle abstract · pdf

  With the increasing usage of machine-learning in high-energy physics analyses, the publication of the trained models in a reusable form has become a crucial question for analysis preservation and reuse. The complexity of these models creates practical issues for both reporting them accurately and for ensuring the stability of their behaviours in different environments and over extended timescales. In this note we discuss the current state of affairs, highlighting specific practical issues and focusing on the most promising technical and strategic approaches to ensure trustworthy analysis-preservation. This material originated from discussions in the LHC Reinterpretation Forum and the 2023 PhysTeV workshop at Les Houches.

  3 citations

• Codebase release 4.0 for RIVET

  Christian Bierlich, Andy Buckley, Jonathan Butterworth, Christian Gutschow, Leif Lonnblad, Tomasz Procter, Peter Richardson, Yoran Yeh

  SciPost Phys. Codebases 36-r4.0 (2024) · published 4 November 2024 | Toggle abstract · src

  The Rivet toolkit is the primary mechanism for phenomenological preservation of collider-physics measurements, containing both a computational core and API for analysis implementation, and a large collection of more than a thousand preserved analyses. In this note we summarise the main changes in the new Rivet 4 major release series. These include a major generalisation and more semantically coherent model for histograms and related data objects, a thorough clean-up of inelegant and legacy observable-computation tools, and new systems for extended analysis-data, incorporation of preserved machine-learning models, and serialization for high-performance computing applications. Where these changes introduce backward-incompatible interface changes, existing analyses have been updated and indications are given on how to update new analysis routines and workflows.

  1 citation

• Robust independent validation of experiment and theory: Rivet version 4 release note

  Christian Bierlich, Andy Buckley, Jonathan Butterworth, Christian Gutschow, Leif Lonnblad, Tomasz Procter, Peter Richardson, Yoran Yeh

  SciPost Phys. Codebases 36 (2024) · published 4 November 2024 | Toggle abstract · pdf

  The Rivet toolkit is the primary mechanism for phenomenological preservation of collider-physics measurements, containing both a computational core and API for analysis implementation, and a large collection of more than a thousand preserved analyses. In this note we summarise the main changes in the new Rivet 4 major release series. These include a major generalisation and more semantically coherent model for histograms and related data objects, a thorough clean-up of inelegant and legacy observable-computation tools, and new systems for extended analysis-data, incorporation of preserved machine-learning models, and serialization for high-performance computing applications. Where these changes introduce backward-incompatible interface changes, existing analyses have been updated and indications are given on how to update new analysis routines and workflows.

  1 citation

• Effective field theory descriptions of Higgs boson pair production

  Lina Alasfar, Luca Cadamuro, Christina Dimitriadi, Arnaud Ferrari, Ramona Gröber, Gudrun Heinrich, Tom Ingebretsen Carlson, Jannis Lang, Serhat Ördek, Laura Pereira Sánchez, Ludovic Scyboz, Jörgen Sjölin

  SciPost Phys. Comm. Rep. 2 (2024) · published 16 October 2024 | Toggle abstract · pdf

  Higgs boson pair production is traditionally considered to be of particular interest for a measurement of the trilinear Higgs self-coupling. Yet it can offer insights into other couplings as well, since -- in an effective field theory (EFT) parameterisation of potential new physics -- both the production cross section and kinematical properties of the Higgs boson pair depend on various other Wilson coefficients of EFT operators. This note summarises the ongoing efforts related to the development of EFT tools for Higgs boson pair production in gluon fusion, and provides recommendations for the use of distinct EFT parameterisations in the Higgs boson pair production process. This document also outlines where further efforts are needed and provides a detailed analysis of theoretical uncertainties. Additionally, benchmark scenarios are updated. We also re-derive a parameterisation of the next-to-leading order (NLO) QCD corrections in terms of the EFT Wilson coefficients both for the total cross section and the distribution in the invariant mass of the Higgs boson pair, providing for the first time also the covariance matrix. A reweighting procedure making use of the newly derived coefficients is validated, which can be used to significantly speed up experimental analyses.

  3 citations

• Higher-form symmetry and chiral transport in real-time Abelian lattice gauge theory

  Arpit Das, Adrien Florio, Nabil Iqbal, Napat Poovuttikul

  SciPost Phys. 17, 085 (2024) · published 19 September 2024 | Toggle abstract · pdf

  We study classical lattice simulations of theories of electrodynamics coupled to charged matter at finite temperature, interpreting them using the higher-form symmetry formulation of magnetohydrodynamics (MHD). We compute transport coefficients using classical Kubo formulas on the lattice and show that the properties of the simulated plasma are in complete agreement with the predictions from effective field theories. In particular, the higher-form formulation allows us to understand from hydrodynamic considerations the relaxation rate of axial charge in the chiral plasma observed in previous simulations. A key point is that the resistivity of the plasma – defined in terms of Kubo formulas for the electric field in the 1-form formulation of MHD – remains a well-defined and predictive quantity at strong electromagnetic coupling. However, the Kubo formulas used to define the conventional conductivity vanish at low frequencies due to electrodynamic fluctuations, and thus the concept of the conductivity of a gauged electric current must be interpreted with care.

  1 citation

• A portable parton-level event generator for the high-luminosity LHC

  Enrico Bothmann, Taylor Childers, Walter Giele, Stefan Höche, Joshua Isaacson, Max Knobbe

  SciPost Phys. 17, 081 (2024) · published 16 September 2024 | Toggle abstract · pdf

  The rapid deployment of computing hardware different from the traditional CPU+RAM model in data centers around the world mandates a change in the design of event generators for the Large Hadron Collider, in order to provide economically and ecologically sustainable simulations for the high-luminosity era of the LHC. Parton-level event generation is one of the most computationally demanding parts of the simulation and is therefore a prime target for improvements. We present a production-ready leading-order parton-level event generation framework capable of utilizing most modern hardware and discuss its performance in the standard candle processes of vector boson and top-quark pair production with up to five additional jets.

  3 citations

• Kicking it off(-shell) with direct diffusion

  Anja Butter, Tomáš Ježo, Michael Klasen, Mathias Kuschick, Sofia Palacios Schweitzer, Tilman Plehn

  SciPost Phys. Core 7, 064 (2024) · published 16 September 2024 | Toggle abstract · pdf

  Off-shell effects in large LHC backgrounds are crucial for precision predictions and, at the same time, challenging to simulate. We present a novel method to transform high-dimensional distributions based on a diffusion neural network and use it to generate a process with off-shell kinematics from the much simpler on-shell one. Applied to a toy example of top pair production at LO we show how our method generates off-shell configurations fast and precisely, while reproducing even challenging on-shell features.

  6 citations

• Anomalies, representations, and self-supervision

  Barry M. Dillon, Luigi Favaro, Friedrich Feiden, Tanmoy Modak, Tilman Plehn

  SciPost Phys. Core 7, 056 (2024) · published 16 August 2024 | Toggle abstract · pdf

  We develop a self-supervised method for density-based anomaly detection using contrastive learning, and test it using event-level anomaly data from CMS ADC2021. The AnomalyCLR technique is data-driven and uses augmentations of the background data to mimic non-Standard-Model events in a model-agnostic way. It uses a permutation-invariant Transformer Encoder architecture to map the objects measured in a collider event to the representation space, where the data augmentations define a representation space which is sensitive to potential anomalous features. An AutoEncoder trained on background representations then computes anomaly scores for a variety of signals in the representation space. With AnomalyCLR we find significant improvements on performance metrics for all signals when compared to the raw data baseline.

  5 citations