Fast Calorimeter Simulation Challenge 2022 - Submissions Dataset 2

These are all the submitted samples to dataset 2 of the “Fast Calorimeter Simulation Challenge 2022”. They each consist of 100k calorimeter showers of electrons with energies sampled from a log-uniform distribution ranging from 1 GeV to 1 TeV.

The training data (based on Geant4) can be found at https://doi.org/10.5281/zenodo.6366271 the paper describing the results is available on arXiv:2410.21611, and further details, in particular helper scripts to parse the data and calculate and visualize basic high-level physics features, are available at https://calochallenge.github.io/homepage/.

The subscripts in the file names corresponds to the individual submissions: 

ID number
Submission name
Original reference

_1
CaloDiffusion
arXiv:2308.03876

_3
conv. L2LFlows
arXiv:2405.20407

_4
CaloINN
arXiv:2312.09290

_5
MDMA
arXiv:2305.15254 arXiv:2408.04997

_7
Calo-VQ
arXiv:2405.06605

_8
CaloScore
arXiv:2206.11898, arXiv:2308.03847

_9
CaloScore distilled
arXiv:2206.11898, arXiv:2308.03847

_10
CaloScore single-shot
arXiv:2206.11898, arXiv:2308.03847

_13
iCaloFlow teacher
arXiv:2305.11934

_14
iCaloFlow student
arXiv:2305.11934

_15
SuperCalo
arXiv:2308.11700

_22
DeepTree
arXiv:2311.12616, arXiv:2312.00042

_23
CaloPointFlow
arXiv:2403.15782

_27
CaloVAE+INN
arXiv:2312.09290

_30
CaloLatent
ML4PS@NeurIPS

_32
CaloDiT
ACAT

_33
CaloDREAM
arXiv:2405.09629

The samples here can be used to reproduce the results of arXiv:2410.21611 and as benchmarks for new models after the challenge concluded.