Frank Gaede
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Frank Gaede has a total of 124 co-authors. Here are the top co-authors by number of shared publications:
- Gregor Kasieczka
Universität Hamburg
8 shared publications - Katja Krüger
Deutsches Elektronen-Synchrotron DESY, Notkestr. 85, 22607 Hamburg, Germany
5 shared publications - Thorsten Buss
Universität Hamburg
5 shared publications - Anatolii Korol
Deutsches Elektronen-Synchrotron DESY, Notkestr. 85, 22607 Hamburg, Germany
4 shared publications - Peter McKeown
CERN, 1211 Geneva 23
4 shared publications Andre Sailer
2 shared publications
Latest Publications
- Open Access
- 13 November 2026
- Published by American Physical Society (APS) in Physical Review D
- Vol. 114 (9)
First full physics benchmark for highly granular calorimeter surrogates
The physics programs of current and future collider experiments necessitate the development of surrogate simulators for calorimeter showers. While much progress has been made in the development of generative models for this task, they have typically been evaluated in simplified scenarios and for single particles. This is particularly true for the challenging task of highly granular calorimeter simulation. For the first time, this work studies the use of highly granular generative calorimeter surrogates in a realistic simulation application. We introduce DDML (4hep with machine learning), a generic library which enables the combination of generative calorimeter surrogates with realistic detectors implemented using the 4hep toolkit. We compare two different generative models—one operating on a regular grid representation and the other using a less common point cloud approach. In order to disentangle methodological details from model performance, we provide comparisons to idealized simulators which directly sample representations of different resolutions from the full simulation ground truth. We then systematically evaluate model performance on post-reconstruction benchmarks for electromagnetic shower simulation. Beginning with a typical single-particle study, we introduce a first multiparticle benchmark based on diphoton separations, before studying a first full physics benchmark based on hadronic decays of the tau lepton. Our results indicate that models operating on a point cloud can achieve a favorable balance between speed and accuracy for highly granular calorimeter simulation compared to those which operate on a regular grid representation.
- 13 November 2026
- Published by American Physical Society (APS) in Physical Review D
- Vol. 114 (9)
Open AccessFirst full physics benchmark for highly granular calorimeter surrogates
The physics programs of current and future collider experiments necessitate the development of surrogate simulators for calorimeter showers. While much progress has been made in the development of generative models for this task, they have typically been evaluated in simplified scenarios and for single particles. This is particularly true for the challenging task of highly granular calorimeter simulation. For the first time, this work studies the use of highly granular generative calorimeter surrogates in a realistic simulation application. We introduce DDML (4hep with machine learning), a generic library which enables the combination of generative calorimeter surrogates with realistic detectors implemented using the 4hep toolkit. We compare two different generative models—one operating on a regular grid representation and the other using a less common point cloud approach. In order to disentangle methodological details from model performance, we provide comparisons to idealized simulators which directly sample representations of different resolutions from the full simulation ground truth. We then systematically evaluate model performance on post-reconstruction benchmarks for electromagnetic shower simulation. Beginning with a typical single-particle study, we introduce a first multiparticle benchmark based on diphoton separations, before studying a first full physics benchmark based on hadronic decays of the tau lepton. Our results indicate that models operating on a point cloud can achieve a favorable balance between speed and accuracy for highly granular calorimeter simulation compared to those which operate on a regular grid representation.
- Open Access
- 1 October 2026
- Published by IOP Publishing in Journal of Instrumentation
- Vol. 21 (10) , P10041
Transferable fast calorimeter shower generation via multi-geometry pre-training
Detailed Geant4 simulation of calorimeter showers dominates the computing budget of high-energy physics experiments. Deep generative surrogates reduce this cost, but they have remained tied to the detector they were trained on, so each new geometry needs a large in-domain dataset. We study whether a single point cloud shower generator can be pre-trained on multiple detectors and transferred to unseen calorimeters. The pre-training geometries come from synthetic geometric variation rather than real-detector data. We introduce SimpleBox , a family of 10 4 box calorimeters spanning the plane of sampling fraction and longitudinal segmentation, and benchmark it against pre-training on realistic detectors. On a calorimeter unseen in pre-training, with 10 3 target showers for fine-tuning, the two priors reduce the aggregated sliced Wasserstein distance to Geant4 by factors of 5.2 (synthetic) and 8.0 (realistic) relative to training from scratch. At larger target sizes the synthetic prior performs better than the realistic one. Geometric diversity alone is therefore a practical way to pre-train a transferable shower generator.
- 1 October 2026
- Published by IOP Publishing in Journal of Instrumentation
- Vol. 21 (10) , P10041
Open AccessTransferable fast calorimeter shower generation via multi-geometry pre-training
Detailed Geant4 simulation of calorimeter showers dominates the computing budget of high-energy physics experiments. Deep generative surrogates reduce this cost, but they have remained tied to the detector they were trained on, so each new geometry needs a large in-domain dataset. We study whether a single point cloud shower generator can be pre-trained on multiple detectors and transferred to unseen calorimeters. The pre-training geometries come from synthetic geometric variation rather than real-detector data. We introduce SimpleBox , a family of 10 4 box calorimeters spanning the plane of sampling fraction and longitudinal segmentation, and benchmark it against pre-training on realistic detectors. On a calorimeter unseen in pre-training, with 10 3 target showers for fine-tuning, the two priors reduce the aggregated sliced Wasserstein distance to Geant4 by factors of 5.2 (synthetic) and 8.0 (realistic) relative to training from scratch. At larger target sizes the synthetic prior performs better than the realistic one. Geometric diversity alone is therefore a practical way to pre-train a transferable shower generator.
- Open Access
- 1 July 2026
- Published by IOP Publishing in Journal of Instrumentation
- Vol. 21 (07) , P07037
Cross-geometry transfer learning in fast electromagnetic shower simulation
Accurate particle shower simulation remains a critical computational bottleneck for high-energy physics. Traditional Monte Carlo methods, such as Geant4 , are computationally prohibitive, while existing machine learning surrogates are tied to specific detector geometries and require complete retraining for each design change or alternative detector. We present a transfer learning methodology for generative calorimeter simulation models that enables adaptation across diverse geometries with high data efficiency. Using point cloud representations and pre-training on the International Large Detector, our approach handles new configurations without re-voxelizing showers for each geometry. On the CaloChallenge dataset, transfer learning with only 100 target-domain samples achieves a 51% improvement on the geometric mean of Wasserstein distance over training from scratch. Parameter-efficient fine-tuning with bias-only adaptation achieves competitive performance while updating only 17% of model parameters. Our analysis provides insight into adaptation mechanisms for particle shower development, establishing a baseline for future progress of point cloud approaches in calorimeter simulation.
- 1 July 2026
- Published by IOP Publishing in Journal of Instrumentation
- Vol. 21 (07) , P07037
Open AccessCross-geometry transfer learning in fast electromagnetic shower simulation
Accurate particle shower simulation remains a critical computational bottleneck for high-energy physics. Traditional Monte Carlo methods, such as Geant4 , are computationally prohibitive, while existing machine learning surrogates are tied to specific detector geometries and require complete retraining for each design change or alternative detector. We present a transfer learning methodology for generative calorimeter simulation models that enables adaptation across diverse geometries with high data efficiency. Using point cloud representations and pre-training on the International Large Detector, our approach handles new configurations without re-voxelizing showers for each geometry. On the CaloChallenge dataset, transfer learning with only 100 target-domain samples achieves a 51% improvement on the geometric mean of Wasserstein distance over training from scratch. Parameter-efficient fine-tuning with bias-only adaptation achieves competitive performance while updating only 17% of model parameters. Our analysis provides insight into adaptation mechanisms for particle shower development, establishing a baseline for future progress of point cloud approaches in calorimeter simulation.
- Top Cited PapersOpen Access
- 1 January 2026
- Published by IOP Publishing in Journal of Instrumentation
- Vol. 21 (01) , P01042
CaloHadronic : a diffusion model for the generation of hadronic showers
Simulating showers of particles in highly-granular calorimeters is a key frontier in the application of machine learning to particle physics. Achieving high accuracy and speed with generative machine learning models can enable them to augment traditional simulations and alleviate a major computing constraint. Recent developments have shown how diffusion based generative shower simulation approaches that do not rely on a fixed structure, but instead generate geometry-independent point clouds, are very efficient. We present a transformer-based extension to previous architectures which were developed for simulating electromagnetic showers in the highly granular electromagnetic calorimeter of the International Large Detector, ILD. The attention mechanism now allows us to generate complex hadronic showers with more pronounced substructure across both the electromagnetic and hadronic calorimeters. This is the first time that machine learning methods are used to holistically generate showers across the electromagnetic and hadronic calorimeter in highly granular imaging calorimeter systems. The code is available at https://github.com/FLC-QU-hep/CaloHadronic .
- 1 January 2026
- Published by IOP Publishing in Journal of Instrumentation
- Vol. 21 (01) , P01042
Top Cited PapersOpen AccessCaloHadronic : a diffusion model for the generation of hadronic showers
Simulating showers of particles in highly-granular calorimeters is a key frontier in the application of machine learning to particle physics. Achieving high accuracy and speed with generative machine learning models can enable them to augment traditional simulations and alleviate a major computing constraint. Recent developments have shown how diffusion based generative shower simulation approaches that do not rely on a fixed structure, but instead generate geometry-independent point clouds, are very efficient. We present a transformer-based extension to previous architectures which were developed for simulating electromagnetic showers in the highly granular electromagnetic calorimeter of the International Large Detector, ILD. The attention mechanism now allows us to generate complex hadronic showers with more pronounced substructure across both the electromagnetic and hadronic calorimeters. This is the first time that machine learning methods are used to holistically generate showers across the electromagnetic and hadronic calorimeter in highly granular imaging calorimeter systems. The code is available at https://github.com/FLC-QU-hep/CaloHadronic .
- Top Cited Papers
- 1 November 2025
- Published by IOP Publishing in Reports on Progress in Physics
- Vol. 88 (11) , 116201
CaloChallenge 2022: a community challenge for fast calorimeter simulation
We present the results of the ‘Fast Calorimeter Simulation Challenge 2022’—the CaloChallenge. We study state-of-the-art generative models on four calorimeter shower datasets of increasing dimensionality, ranging from a few hundred voxels to a few tens of thousand voxels. The 31 individual submissions span a wide range of current popular generative architectures, including variational autoencoders (VAEs), generative adversarial networks (GANs), normalizing flows, diffusion models, and models based on conditional flow matching. We compare all submissions in terms of quality of generated calorimeter showers, as well as shower generation time and model size. To assess the quality we use a broad range of different metrics including differences in one-dimensional histograms of observables, KPD/FPD scores, AUCs of binary classifiers, and the log-posterior of a multiclass classifier. The results of the CaloChallenge provide the most complete and comprehensive survey of cutting-edge approaches to calorimeter fast simulation to date. In addition, our work provides a uniquely detailed perspective on the important problem of how to evaluate generative models. As such, the results presented here should be applicable for other domains that use generative AI and require fast and faithful generation of samples in a large phase space. Report Numbers : HEPHY-ML-24-05, FERMILAB-PUB-24-0728-CMS, TTK-24-43.
- 1 November 2025
- Published by IOP Publishing in Reports on Progress in Physics
- Vol. 88 (11) , 116201
Top Cited PapersCaloChallenge 2022: a community challenge for fast calorimeter simulation
We present the results of the ‘Fast Calorimeter Simulation Challenge 2022’—the CaloChallenge. We study state-of-the-art generative models on four calorimeter shower datasets of increasing dimensionality, ranging from a few hundred voxels to a few tens of thousand voxels. The 31 individual submissions span a wide range of current popular generative architectures, including variational autoencoders (VAEs), generative adversarial networks (GANs), normalizing flows, diffusion models, and models based on conditional flow matching. We compare all submissions in terms of quality of generated calorimeter showers, as well as shower generation time and model size. To assess the quality we use a broad range of different metrics including differences in one-dimensional histograms of observables, KPD/FPD scores, AUCs of binary classifiers, and the log-posterior of a multiclass classifier. The results of the CaloChallenge provide the most complete and comprehensive survey of cutting-edge approaches to calorimeter fast simulation to date. In addition, our work provides a uniquely detailed perspective on the important problem of how to evaluate generative models. As such, the results presented here should be applicable for other domains that use generative AI and require fast and faithful generation of samples in a large phase space. Report Numbers : HEPHY-ML-24-05, FERMILAB-PUB-24-0728-CMS, TTK-24-43.
- Open Access
- 1 July 2025
- Published by IOP Publishing in Journal of Instrumentation
- Vol. 20 (07) , P07007
OmniJet-α_C: learning point cloud calorimeter simulations using generative transformers
We show the first use of generative transformers for generating calorimeter showers as point clouds in a high-granularity calorimeter. Using the tokenizer and generative part of the OmniJet-α model, we represent the hits in the detector as sequences of integers. This model allows variable-length sequences, which means that it supports realistic shower development and does not need to be conditioned on the number of hits. Since the tokenization represents the showers as point clouds, the model learns the geometry of the showers without being restricted to any particular voxel grid.
- 1 July 2025
- Published by IOP Publishing in Journal of Instrumentation
- Vol. 20 (07) , P07007
Open AccessOmniJet-α_C: learning point cloud calorimeter simulations using generative transformers
We show the first use of generative transformers for generating calorimeter showers as point clouds in a high-granularity calorimeter. Using the tokenizer and generative part of the OmniJet-α model, we represent the hits in the detector as sequences of integers. This model allows variable-length sequences, which means that it supports realistic shower development and does not need to be conditioned on the number of hits. Since the tokenization represents the showers as point clouds, the model learns the geometry of the showers without being restricted to any particular voxel grid.
- Open Access
- 1 September 2024
- Published by IOP Publishing in Journal of Instrumentation
- Vol. 19 (09) , P09003
Convolutional L2LFlows: generating accurate showers in highly granular calorimeters using convolutional normalizing flows
In the quest to build generative surrogate models as computationally efficient alternatives to rule-based simulations, the quality of the generated samples remains a crucial frontier. So far, normalizing flows have been among the models with the best fidelity. However, as the latent space in such models is required to have the same dimensionality as the data space, scaling up normalizing flows to high dimensional datasets is not straightforward. The prior L2LFlows approach successfully used a series of separate normalizing flows and sequence of conditioning steps to circumvent this problem. In this work, we extend L2LFlows to simulate showers with a 9-times larger profile in the lateral direction. To achieve this, we introduce convolutional layers and U-Net-type connections, move from masked autoregressive flows to coupling layers, and demonstrate the successful modelling of showers in the ILD Electromagnetic Calorimeter as well as Dataset 3 from the public CaloChallenge dataset.
- 1 September 2024
- Published by IOP Publishing in Journal of Instrumentation
- Vol. 19 (09) , P09003
Open AccessConvolutional L2LFlows: generating accurate showers in highly granular calorimeters using convolutional normalizing flows
In the quest to build generative surrogate models as computationally efficient alternatives to rule-based simulations, the quality of the generated samples remains a crucial frontier. So far, normalizing flows have been among the models with the best fidelity. However, as the latent space in such models is required to have the same dimensionality as the data space, scaling up normalizing flows to high dimensional datasets is not straightforward. The prior L2LFlows approach successfully used a series of separate normalizing flows and sequence of conditioning steps to circumvent this problem. In this work, we extend L2LFlows to simulate showers with a 9-times larger profile in the lateral direction. To achieve this, we introduce convolutional layers and U-Net-type connections, move from masked autoregressive flows to coupling layers, and demonstrate the successful modelling of showers in the ILD Electromagnetic Calorimeter as well as Dataset 3 from the public CaloChallenge dataset.
- Open Access
- 19 January 2024
- Published by Sissa Medialab Srl
- Vol. 449, 548
Development of the time-of-flight particle identification for future Higgs factories
With the emergence of advanced Silicon (Si) sensor technologies such as LGADs, it is now possible to achieve exceptional time measurement precision below 50 ps. As a result, the implementation of time-of-flight (TOF) particle identification (PID) for charged hadrons at future $e^{+}e^{-}$ Higgs factory detectors has gained increasing attention. Other PID techniques require a gaseous tracker with excellent dE/dx resolution, or a Ring-imaging Cherenkov detector (RICH), which adds additional material in front of the calorimeter. TOF measurements can be implemented either in the outer layers of the tracker or in the electromagnetic calorimeter, and are thus particularly interesting as a PID method for detector concepts based on all-silicon trackers and optimised for particle-flow reconstruction. In this study, we will explore potential integration scenarios of a TOF measurement in a future Higgs factory detector, using the International Large Detector (ILD) as an example. We will focus on the challenges associated with crucial components of TOF PID, namely track length reconstruction and TOF measurements. The subsequent discussion will highlight the vital impact of precise track length reconstruction and various TOF measurement techniques, including recently developed machine learning approaches. We will evaluate the performance in terms of $\pi/K$ and $K/p$ separation as a function of momentum, and discuss potential physics applications.- 19 January 2024
- Published by Sissa Medialab Srl
- Vol. 449, 548
Open AccessDevelopment of the time-of-flight particle identification for future Higgs factories
With the emergence of advanced Silicon (Si) sensor technologies such as LGADs, it is now possible to achieve exceptional time measurement precision below 50 ps. As a result, the implementation of time-of-flight (TOF) particle identification (PID) for charged hadrons at future $e^{+}e^{-}$ Higgs factory detectors has gained increasing attention. Other PID techniques require a gaseous tracker with excellent dE/dx resolution, or a Ring-imaging Cherenkov detector (RICH), which adds additional material in front of the calorimeter. TOF measurements can be implemented either in the outer layers of the tracker or in the electromagnetic calorimeter, and are thus particularly interesting as a PID method for detector concepts based on all-silicon trackers and optimised for particle-flow reconstruction. In this study, we will explore potential integration scenarios of a TOF measurement in a future Higgs factory detector, using the International Large Detector (ILD) as an example. We will focus on the challenges associated with crucial components of TOF PID, namely track length reconstruction and TOF measurements. The subsequent discussion will highlight the vital impact of precise track length reconstruction and various TOF measurement techniques, including recently developed machine learning approaches. We will evaluate the performance in terms of $\pi/K$ and $K/p$ separation as a function of momentum, and discuss potential physics applications. - Open Access
Development of time-of-flight particle identification for future Higgs factories
With the emergence of advanced Silicon (Si) sensor technologies such as LGADs, it is now possible to achieve exceptional time measurement precision below 50 ps. As a result, the implementation of time-of-flight (TOF) particle identification (PID) for charged hadrons at future $e^{+}e^{-}$ Higgs factory detectors has gained increasing attention. Other PID techniques require a gaseous tracker with excellent dE/dx resolution, or a Ring-imaging Cherenkov detector (RICH), which adds additional material in front of the calorimeter. TOF measurements can be implemented either in the outer layers of the tracker or in the electromagnetic calorimeter, and are thus particularly interesting as a PID method for detector concepts based on all-silicon trackers and optimised for particle-flow reconstruction. In this study, we will explore potential integration scenarios of a TOF measurement in a future Higgs factory detector, using the International Large Detector (ILD) as an example. We will focus on the challenges associated with crucial components of TOF PID, namely track length reconstruction and TOF measurements. The subsequent discussion will highlight the vital impact of precise track length reconstruction and various TOF measurement techniques, including recently developed machine learning approaches. We will evaluate the performance in terms of $π/K$ and $K/p$ separation as a function of momentum, and discuss potential physics applications.Open AccessDevelopment of time-of-flight particle identification for future Higgs factories
With the emergence of advanced Silicon (Si) sensor technologies such as LGADs, it is now possible to achieve exceptional time measurement precision below 50 ps. As a result, the implementation of time-of-flight (TOF) particle identification (PID) for charged hadrons at future $e^{+}e^{-}$ Higgs factory detectors has gained increasing attention. Other PID techniques require a gaseous tracker with excellent dE/dx resolution, or a Ring-imaging Cherenkov detector (RICH), which adds additional material in front of the calorimeter. TOF measurements can be implemented either in the outer layers of the tracker or in the electromagnetic calorimeter, and are thus particularly interesting as a PID method for detector concepts based on all-silicon trackers and optimised for particle-flow reconstruction. In this study, we will explore potential integration scenarios of a TOF measurement in a future Higgs factory detector, using the International Large Detector (ILD) as an example. We will focus on the challenges associated with crucial components of TOF PID, namely track length reconstruction and TOF measurements. The subsequent discussion will highlight the vital impact of precise track length reconstruction and various TOF measurement techniques, including recently developed machine learning approaches. We will evaluate the performance in terms of $π/K$ and $K/p$ separation as a function of momentum, and discuss potential physics applications. - Open Access
- 26 November 2022
- Published by Sissa Medialab Srl
- Vol. 414, 1237
EDM4hep - a common event data model for HEP experiments
A shared, common event data model, EDM4hep, is an integral part of the Key4hep project. EDM4hep aims to be usable by all future collider projects, despite their different collision environments and the different detector technologies that are under discussion. This constitutes a major challenge that EDM4hep addresses by using podio, a C++ toolkit for the creation and handling of event data models, developed in the context of the AIDA R&D program. This approach allows for quick prototyping of new data types and provides a streamlined framework for updates. After presenting an overview of the basic features of EDM4hep and podio, we will discuss the current experience with an initial version of EDM4hep in different physics studies. Additionally, we will present the planned developments that are necessary for a first stable version of EDM4hep, addressing in particular backward compatibility aspects and schema evolution. We will conclude with an outlook on the future developments directions beyond this first stable version.
- 26 November 2022
- Published by Sissa Medialab Srl
- Vol. 414, 1237
Open AccessEDM4hep - a common event data model for HEP experiments
A shared, common event data model, EDM4hep, is an integral part of the Key4hep project. EDM4hep aims to be usable by all future collider projects, despite their different collision environments and the different detector technologies that are under discussion. This constitutes a major challenge that EDM4hep addresses by using podio, a C++ toolkit for the creation and handling of event data models, developed in the context of the AIDA R&D program. This approach allows for quick prototyping of new data types and provides a streamlined framework for updates. After presenting an overview of the basic features of EDM4hep and podio, we will discuss the current experience with an initial version of EDM4hep in different physics studies. Additionally, we will present the planned developments that are necessary for a first stable version of EDM4hep, addressing in particular backward compatibility aspects and schema evolution. We will conclude with an outlook on the future developments directions beyond this first stable version.