ACAT 2025

Session

Track 2: Data Analysis - Algorithms and Tools

8 Sept 2025, 14:30

ESA B

Conveners

Track 2: Data Analysis - Algorithms and Tools

• co-chair: Daniel Murnane
• chair: Frank Gaede

Track 2: Data Analysis - Algorithms and Tools

• chair: Davide Valsecchi
• co-chair: Daniel Murnane

Track 2: Data Analysis - Algorithms and Tools

• chair: Luisa Lucie-Smith
• co-chair: Louis Moureaux

Track 2: Data Analysis - Algorithms and Tools

• chair: Frank Gaede
• co-chair: Luisa Lucie-Smith

Track 2: Data Analysis - Algorithms and Tools

• chair: Tilman Plehn
• co-chair: Karim El Morabit

Track 2: Data Analysis - Algorithms and Tools

• co-chair: David Rousseau
• chair: Thea Aarrestad

Track 2: Data Analysis - Algorithms and Tools

• chair: Thea Aarrestad
• co-chair: Tilman Plehn

72. Developments of GNN Track Reconstruction for the ATLAS ITk Detector

Jay Chan (Lawrence Berkeley National Lab. (US))

08/09/2025, 14:30

Track 2: Data Analysis - Algorithms and Tools

Oral

Track reconstruction is a cornerstone of modern collider experiments, and the HL-LHC ITk upgrade for ATLAS poses new challenges with its increased silicon hit clusters and strict throughput requirements. Deep learning approaches compare favorably with traditional combinatorial ones — as shown by the GNN4ITk project, a geometric learning tracking pipeline that achieves competitive physics...

89. CyberPFA: Particle Flow Algorithm for Crystal Bar ECAL

Yang Zhang (Institute of High Energy Physics, Chinese Academy of Science)

08/09/2025, 14:50

Track 2: Data Analysis - Algorithms and Tools

Oral

Precision measurements of Higgs, W, and Z bosons at future lepton colliders demand jet energy reconstruction with unprecedented accuracy. The particle flow approach has proven to be an effective method for achieving the required jet energy resolution. We present CyberPFA, a particle flow algorithm specifically optimized for the particle-flow-oriented crystal bar electromagnetic calorimeter...

90. Attention-Enhanced Lightweight GNNs for LHCb Next-generation Particle reconstruction and Identification

Cilicia Uzziel Perez (La Salle, Ramon Llull University (ES))

08/09/2025, 15:10

Track 2: Data Analysis - Algorithms and Tools

Oral

We present lightweight, attention-enhanced Graph Neural Networks (GNNs) tailored for real-time particle reconstruction and identification in LHCb’s next-generation calorimeter. Our architecture builds on node-centric GarNet layers, which eliminate costly edge message passing and are optimized for FPGA deployment, achieving sub-microsecond inference latency. By integrating attention mechanisms...

137. GNN-based E2E reconstruction in different highly granular calorimeters

Katharina sophia Schaeuble, Ulrich Einhaus (KIT - Karlsruhe Institute of Technology (DE))

08/09/2025, 15:30

Track 2: Data Analysis - Algorithms and Tools

Oral

We present a versatile GNN-based end-to-end reconstruction algorithm for highly granular calorimeters that can include track and timing information to aid the reconstruction of particles. The algorithm starts directly from calorimeter hits and possibly reconstructed tracks, and outputs a coordinate transformation in which all shower objects are well separated from each other and assigned...

148. TMVA SOFIE: Enhancements in ML Inference through graph optimizations and heterogeneous architectures

Enrico Lupi (CERN, INFN Padova (IT))

08/09/2025, 15:50

Track 2: Data Analysis - Algorithms and Tools

Oral

With the upcoming High-Luminosity upgrades at the LHC, data generation rates are expected to increase significantly. This calls for highly efficient architectures for machine learning inference in experimental workflows like event reconstruction, simulation, and data analysis.
At the ML4EP team at CERN, we have developed SOFIE, a tool within the ROOT/TMVA package that translates externally...

243. On focusing statistical power for searches and measurements in particle physics

Aishik Ghosh (University of California Irvine (US))

08/09/2025, 16:40

Track 2: Data Analysis - Algorithms and Tools

Oral

Particle physics experiments rely on the (generalised) likelihood ratio test (LRT) for searches and measurements. This is not guaranteed to be optimal for composite hypothesis tests, as the Neyman-Pearson lemma pertains only to simple hypothesis tests. An improvement in the core statistical testing methodology would have widespread ramifications across experiments. We discuss an alternate test...

194. Parameter Estimation with Neural Simulation-Based Inference in ATLAS

R D Schaffer (Université Paris-Saclay (FR))

08/09/2025, 17:00

Track 2: Data Analysis - Algorithms and Tools

Oral

Neural Simulation-Based Inference (NSBI) is a powerful class of machine learning (ML)-based methods for statistical inference that naturally handle high dimensional parameter estimation without the need to bin data into low-dimensional summary histograms. Such methods are promising for a range of measurements at the Large Hadron Collider, where no single observable may be optimal to scan over...

223. Modular Data-Driven Calibration and Analysis Correction in ALICE

Marian I Ivanov (GSI - Helmholtzzentrum fur Schwerionenforschung GmbH (DE))

08/09/2025, 17:20

Track 2: Data Analysis - Algorithms and Tools

Oral

We present a modular, data-driven framework for calibration and performance correction in the ALICE experiment. The method addresses time- and parameter-dependent effects in high-occupancy heavy-ion environments, where evolving detector conditions (e.g., occupancy and cluster overlaps, gain drift, space charge, dynamic distortions, and reconstruction or calibration deficiencies) require...

163. Machine learning applications in the JUNO experiment

Hongyue Duyang (Shandong University)

08/09/2025, 17:40

Track 2: Data Analysis - Algorithms and Tools

Oral

Jiangmen Underground Neutrino Observatory (JUNO) is a next generation 20-kton liquid scintillator detector under construction in southern China. It is designed to determine neutrino mass ordering via the measurement of reactor neutrino oscillation, and also to study other physics topics including atmospheric neutrinos, supernova neutrinos and more. The detector's large mass and high...

11. Reconstructing tau leptons with a cross-task, cross-detector foundation model

Laurits Tani (National Institute of Chemical Physics and Biophysics (EE))

09/09/2025, 14:30

Track 2: Data Analysis - Algorithms and Tools

Oral

The application of foundation models in high-energy physics has recently been proposed as a way to use large unlabeled datasets to efficiently train powerful task-specific models. The aim is to train a task-agnostic model on an existing large dataset such that the learned representation can later be utilized for subsequent downstream physics tasks.
The pretrained model can reduce the training...

87. OmniJet-alpha: foundation model updates

Anna Hallin (University of Hamburg)

09/09/2025, 14:50

Track 2: Data Analysis - Algorithms and Tools

Oral

OmniJet-alpha, the first cross-task foundation model for particle physics, was first presented at ACAT 2024. In its base configuration, OmniJet-alpha is capable of transfer learning between an unsupervised problem (jet generation) and a classic supervised task (jet tagging). Since its release, we have also shown that it can sucessfully transfer from CMS Open data to simulation, and even...

70. Multi-Modal track reconstruction using Graph Neural Networks at Belle II

Lea Reuter (Karlsruhe Institute of Technology)

09/09/2025, 15:10

Track 2: Data Analysis - Algorithms and Tools

Oral

Large backgrounds and detector aging impact the track finding in the Belle II central drift chamber, reducing both purity and efficiency in events. This necessitates the development of new track algorithms to mitigate detector performance degradation. Building on our previous success with an end-to-end multi-track reconstruction algorithm for the Belle II experiment at the SuperKEKB collider...

66. CMS FlashSim: end-to-end simulation with ML

CMS Collaboration, Filippo Cattafesta (Scuola Normale Superiore & INFN Pisa (IT))

09/09/2025, 15:30

Track 2: Data Analysis - Algorithms and Tools

Oral

Detailed event simulation at the LHC is taking a large fraction of computing budget. CMS developed an end-to-end ML based simulation that can speed up the time for production of analysis samples of several orders of magnitude with a limited loss of accuracy. As the CMS experiment is adopting a common analysis level format, the NANOAOD, for a larger number of analyses, such an event...

256. Computing the Matrix Element Method with generative machine learning

CMS Collaboration, Dr Florian Bury (University of Bristol)

09/09/2025, 15:50

Track 2: Data Analysis - Algorithms and Tools

Oral

The Matrix Element Method (MEM) offers optimal statistical power for hypothesis testing in particle physics, but its application is hindered by the computationally intensive multi-dimensional integrals required to model detector effects. We present a novel approach that addresses this challenge by employing Transformers and generative machine learning (ML) models. Specifically, we utilize ML...

145. Interaction-Aware and Domain-Invariant Representation Learning for Inclusive Flavour Tagging

Quentin Führing (Technische Universitaet Dortmund (DE), University of Cambridge (UK))

09/09/2025, 16:40

Track 2: Data Analysis - Algorithms and Tools

Oral

Measurements of neutral, oscillating mesons are a gateway to quantum mechanics and give access to the fundamental interactions of elementary particles. For example, precise measurements of $CP$ violation in neutral $B$ mesons can be taken in order to test the Standard Model of particle physics. These measurements require knowledge of the $B$-meson flavour at the time of its production, which...

149. How to make any Network Lorentz-Equivariant

Sebastian Pitz (ITP, Heidelberg University)

09/09/2025, 17:00

Track 2: Data Analysis - Algorithms and Tools

Oral

We construct Lorentz-equivariant transformer and graph networks using the concept of local canonicalization. While many Lorentz-equivariant architectures use specialized layers, this approach allows to take any existing non-equivariant architecture and make it Lorentz-equivariant using transformations with equivariantly predicted local frames. In addition, data augmentation emerges as a...

156. Hyperparameter Transfer for Graph Transformers

Gage DeZoort (Princeton University (US))

09/09/2025, 17:20

Track 2: Data Analysis - Algorithms and Tools

Oral

Modern machine learning (ML) algorithms are sensitive to the specification of non-trainable parameters called hyperparameters (e.g., learning rate or weight decay). Without guiding principles, hyperparameter optimization is the computationally expensive process of sweeping over various model sizes and, at each, re-training the model over a grid of hyperparameter settings. However, recent...

171. Evaluating Two-Sample Tests for Validating Generators in Precision Sciences

Samuele Grossi (Università degli studi di Genova & INFN sezione di Genova)

09/09/2025, 17:40

Track 2: Data Analysis - Algorithms and Tools

Oral

Deep generative models have become powerful tools for alleviating the computational burden of traditional Monte Carlo generators in producing high-dimensional synthetic data. However, validating these models remains challenging, especially in scientific domains requiring high precision, such as particle physics. Two-sample hypothesis testing offers a principled framework to address this task....

3. Real-Time event reconstruction for Nuclear Physics Experiments using Artificial Intelligence

Gagik Gavalian (Jefferson National Lab)

10/09/2025, 11:30

Track 2: Data Analysis - Algorithms and Tools

Oral

Charged track reconstruction is a critical task in nuclear physics experiments, enabling the identification and analysis of particles produced in high-energy collisions. Machine learning (ML) has emerged as a powerful tool for this purpose, addressing the challenges posed by complex detector geometries, high event multiplicities, and noisy data. Traditional methods rely on pattern recognition...

51. Machine Learning algorithms for the COSI mission background rejection

Francesco Fenu (Agenzia Spaziale Italiana)

10/09/2025, 11:50

Track 2: Data Analysis - Algorithms and Tools

Oral

The Compton Spectrometer and Imager (COSI) is a NASA Small Explorer (SMEX) satellite mission planned to fly in 2027. It has the participation of institutions in the US, Europe and Asia and aims at the construction of a gamma-ray telescope for observations in the 0.2-5 MeV energy range. COSI consists of an array of germanium strip detectors cooled to cryogenic temperatures with millimeter...

76. Probing The Invisible Solar System Through Stellar Eclipse Events

Dr Marcel Völschow (Hamburg University of Applied Sciences)

10/09/2025, 12:10

Track 2: Data Analysis - Algorithms and Tools

Oral

Beyond the planet Neptune, only the largest solar system objects can be observed directly. However, there are tens of thousands of smaller objects whose frequency and distribution could provide valuable insights into the formation of our solar system - if we could see them.

Project SOWA (Solar-system Occultation Watch and Analysis) aims to systematically search for such invisible objects...

189. AI Agents for Ground-Based Gamma Astronomy

Julian Simon Schliwinski (Humboldt University of Berlin (DE))

10/09/2025, 12:30

Track 2: Data Analysis - Algorithms and Tools

Oral

The next generation of ground-based gamma-ray astronomy instruments will involve arrays of dozens of telescopes, leading to an increase in operational and analytical complexity. This scale-up poses challenges for both system operations and offline data processing, especially when conventional approaches struggle to scale effectively. To address these challenges, we are developing AI agents...

202. Neural Quasiprobabilistic Likelihood Ratio Estimation with Negatively Weighted Data

Stephen Jiggins (Deutsches Elektronen-Synchrotron (DE))

10/09/2025, 12:50

Track 2: Data Analysis - Algorithms and Tools

Oral

In many domains of science the likelihood function is a fundamental ingredient used to statistically infer model parameters from data, due to the likelihood ratio (LR) as an optimal test statistic. Neural based LR estimation using probabilistic classification has therefore had a significant impact in these domains, providing a scalable method for determining an intractable LR from simulated...

190. Upgrade of the Belle II First-Level Neural Track Trigger by Three-Dimensional Hough Finding and Deep Neural Networks on FPGAs

Simon Hiesl

11/09/2025, 14:30

Track 2: Data Analysis - Algorithms and Tools

Oral

In anticipation of higher luminosities at the Belle II experiment, high levels of beam background
from outside of the interaction region are expected. To prevent track trigger rates
from surpassing the limitations of the data acquisition system, an upgrade of the first-level
neural track trigger becomes indispensable. This upgrade contains a novel track finding
algorithm based on...

220. Track fitting at the full LHC collision frequency: Design and performance of the GPU-based Kalman Filter at the LHCb experiment

Lennart Uecker (Heidelberg University (DE))

11/09/2025, 14:50

Track 2: Data Analysis - Algorithms and Tools

Oral

The LHCb experiment at the Large Hadron Collider (LHC) operates a fully software-based trigger system that processes proton-proton collisions at a rate of 30 MHz, reconstructing both charged and neutral particles in real time. The first stage of this trigger system, running on approximately 500 GPU cards, performs a track pattern recognition to reconstruct particle trajectories with low...

200. Real-time monitoring of LHCb beam spot properties based on FPGA hit reconstruction

Giulio Cordova (Universita & INFN Pisa (IT))

11/09/2025, 15:10

Track 2: Data Analysis - Algorithms and Tools

Oral

The upgraded LHCb experiment is pioneering the landscape of real-time data-processing techniques using an heterogeneous computing infrastructure, composed of both GPUs and FPGAs, aimed at boosting the performance of the HLT1 reconstruction. Amongst the novelties in the reconstruction infrastructure made for the Run 3, the introduction of a real-time VELO hit-finding FPGA-based architecture...

53. Improving the CMS High Level Trigger tracking at the HL-LHC with novel and evolved heterogeneous algorithms

CMS Collaboration, Mario Masciovecchio (Univ. of California San Diego (US))

11/09/2025, 15:30

Track 2: Data Analysis - Algorithms and Tools

Oral

Charged particle track reconstruction is one the heaviest computational tasks in the event reconstruction chain at Large Hadron Collider (LHC) experiments. Furthermore, projections for the High Luminosity LHC (HL-LHC) show that the required computing resources for single-threaded CPU algorithms will exceed those that are expected to be available. It follows that experiments at the HL-LHC will...

161. Deep Learning for Primary Vertex Identification in the ATLAS Experiment

Qi Bin Lei (Stanford University (US)), Rocky Bala Garg (Stanford University (US))

11/09/2025, 15:50

Track 2: Data Analysis - Algorithms and Tools

Oral

The exponential time scaling of traditional primary vertex reconstruction algorithms raises significant performance concerns for future high-pileup environments, particularly with the upcoming High Luminosity upgrade to the Large Hadron Collider. In this talk, we introduce PV-Finder, a deep learning-based approach that leverages reconstructed track parameters to directly predict primary vertex...

139. Generative Unfolding in Many Dimensions

Antoine Petitjean (ITP, Universität Heidelberg)

11/09/2025, 16:40

Track 2: Data Analysis - Algorithms and Tools

Oral

Unfolding detector-level data into meaningful particle-level distributions remains a key challenge in collider physics, especially as the dimensionality of the relevant observables increases. Traditional unfolding techniques often struggle with such high-dimensional problems, motivating the development of machine learning-based approaches.We introduce a new method for generative unfolding that...

152. Full Generative Unfolding

Sofia Palacios Schweitzer (ITP, University Heidelberg)

11/09/2025, 17:00

Track 2: Data Analysis - Algorithms and Tools

Oral

Two shortcomings of classical unfolding algorithms, namely that they are defined on binned, one-dimensional observables, can be overcome when using generative machine learning. Many studies on generative unfolding reduce the problem to correcting for detector smearing, however a full unfolding pipeline must also account for background, acceptance and efficiency effects. To fully integrate...

169. Bridging the Gap Between Unfolding and Quantification Learning

Dr Mirko Bunse (Lamarr Institute for Machine Learning and Artificial Intelligence, Dortmund, Germany)

11/09/2025, 17:20

Track 2: Data Analysis - Algorithms and Tools

Oral

Measured distributions are usually distorted by a finite resolution of the detector. Within physics research, the necessary correction of these distortions is know as Unfolding. Machine learning research uses a different term for this very task: Quantification Learning. For the past two decades, this difference in terminology - together with several differences in notation - have prevented...

158. Efficient bin by bin profile likelihood minimization for precision measurements

David Walter (Massachusetts Inst. of Technology (US))

11/09/2025, 17:40

Track 2: Data Analysis - Algorithms and Tools

Oral

The High-Luminosity LHC era will deliver unprecedented data volumes, enabling measurements on fine-grained multidimensional histograms containing millions of bins with thousands of events each. Achieving ultimate precision requires modeling thousands of systematic uncertainty sources, creating computational challenges for likelihood minimization and parameter extraction. Fast minimization is...