TPCpp-10M : Simulated proton-proton collisions in a Time Projection Chamber for AI Foundation Models Journal: Data in Brief
Shuhang Li, Yi Huang, David Park, Xihaier Luo, Haiwang Yu, Yeonju Go, Christopher Pinkenburg, Yuewei Lin, Shinjae Yoo, Joseph Osborn, Christof Roland, Jin Huang, Yihui Ren · 2025
Read TPCpp-10M : Simulated proton-proton collisions in a Time Projection Chamber for AI Foundation Models Journal: Data in Brief by Shuhang Li, Yi Huang, David Park, Xihaier Luo, Haiwang Yu, Yeonju Go, Christopher Pinkenburg, Yuewei Lin, Shinjae Yoo, Joseph Osborn, Christof Roland, Jin Huang, Yihui Ren (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-2509-05792.txt.
Introduction
A shared dataset for training AI foundation models on particle collisions is introduced in a 2025 paper led by Shuhang Li with twelve co-authors. The set holds ten million simulated proton-proton collisions, recorded as if by the sPHENIX detector's Time Projection Chamber at Brookhaven. Particle physics lacks open data at this scale, which slows joint work with machine learning researchers.
Events were generated with the Pythia simulator and passed through Geant4 to mimic real detector signals and noise. Alongside the unlabeled events sit seventy thousand labeled examples for three tasks: finding tracks, identifying particles, and tagging noise. Files come in a common NumPy format so outsiders can start without specialist software. The release gives both fields a common testbed for studying scaling and transfer.
Catalog record
| CREATOR | Shuhang Li, Yi Huang, David Park, Xihaier Luo, Haiwang Yu, Yeonju Go, Christopher Pinkenburg, Yuewei Lin, Shinjae Yoo, Joseph Osborn, Christof Roland, Jin Huang, Yihui Ren |
|---|---|
| YEAR | 2025 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 16 KB |
| VERSION | 1.0 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2509-05792
Browse the full library · About the Tower of Babel
Full text
The opening of the work, from the library's hosted text:
TPCpp-10M : Simulated proton-proton collisions in a Time Projection Chamber for AI Foundation Models Journal: Data in Brief Shuhang Li, Yi Huang, David Park, Xihaier Luo, Haiwang Yu, Yeonju Go, Christopher Pinkenburg, Yuewei Lin, Shinjae Yoo, Joseph Osborn, Christof Roland, Jin Huang, Yihui Ren Abstract Scientific foundation models hold great promise for advancing nuclear and particle physics by improving analysis precision and accelerating discovery. Yet, progress in this field is often limited by the lack of openly available large-scale datasets, as well as standardized evaluation tasks and metrics. Furthermore, the specialized knowledge and software typically required to process particle physics data pose significant barriers to interdisciplinary collaboration with the broader machine learning community. This work introduces a large, openly accessible dataset of 10 million simulated proton–proton collisions, designed to support self-supervised training of foundation models. To facilitate ease of use, the dataset is provided in a common NumPy format. In addition, it includes 70,000 labeled examples spanning three well-defined downstream tasks – track finding, particle identification, and noise tagging – to enable systematic evaluation of the foundation model’s adaptability. The simulated data are generated using the Pythia Monte Carlo event generator at a center-of-mass energy of s = 200 GeV \sqrt{s}=$200\text{\,}\mathrm{GeV}$ and processed with Geant4 to include realistic detector conditions and signal emulation in the sPHENIX Time Projection Chamber at the Relativistic Heavy Ion Collider, located at Brookhaven National Laboratory. This dataset resource establishes a common ground for interdisciplinary research, enabling machine learning scientists and physicists alike to explore scaling behaviors, assess transferability, and accelerate progress toward foundation models in nuclear and high-energy physics. The complete simulation and reconstruction chain is reproducible with the sPHENIX software stack. All data and code locations are provided under Data Accessibility. Specifications Table Subject Physics Specific subject area Charged-particle tracking simulation in a time projection chamber (TPC) at Relativistic Heavy Ion Collider (RHIC) Type of data & formats NumPy compressed archive Data collection proton+proton ( p{+}p ) collision events generated with Pythia 8.307 (Detroit tune) [ 1 , 2 ] at center-of-mass energy \sqrt{s}=$200\text{\,}\mathrm{GeV}$ ; simulated with Geant4 [ 3 ] (FTFP_BERT_HP) via the as-built sPHENIX detector and 1.4\text{\,}\mathrm{T} magnetic field; TPC signals digitized and reconstructed into spacepoints using sPHENIX software. Labeled set includes spacepoint-to-truth associations; Data source location Brookhaven National Laboratory, Upton, NY, USA (simulated under sPHENIX software environment). Data accessibility Repository name: Zenodo Data identification number (DOI): 10.5281/zenodo.16970029 Direct URL to data: https://doi.org/10.5281/zenodo.16970029 Instructions for access: Public, anonymous download; dataset provided as compressed archives; example loaders and documentation included in the repository. Related research article None Figure 1: RHIC and sPHENIX experiment (context figure). Value of the Data 1. Supports foundation model training. The dataset provides 10,000,000 unlabeled events suitable for large-scale self-supervised or contrastive pretraining, complemented by 70,000 labeled events for fine-tuning. This enables scalable applications of machine learning to charged-particle tracking. 2. Fills a gap in open high-energy physics datasets. Most publicly available tracking datasets are based on silicon detectors (e.g., strip or pixel trackers) in high-energy particle collisions. In contrast, this dataset is built on a large, gas-filled TPC, the central tracker in sPHENIX and many other nuclear physics experiments. By providing realistic TPC spacepoints with truth associations, it enables research on tracking challenges unique to gaseous detectors, such as long drift distances, diffusion, and complex pattern recognition across several tens of layers, compared to the \sim 10 precise two-dimensional layers typically found in silicon trackers. 3. Benchmarking resource. The labeled subset includes ground-truth information for three standard downstream tasks – track finding, particle identification, and noise tagging – along with recommended metrics, establishing a common benchmark for fair and reproducible evaluation. Background In nuclear and particle physics experiments, tracking detectors are used to measure and reconstruct the trajectories of charged particles. These measurements are crucial for determining momenta, identifying particle species, and studying underlying physics processes. The TPC provides three-dimensional (3D) spacepoints that reconstruct the trajectories of particles traversing the detector. Figure 1 shows the sPHENIX TPC at RHIC is a cylindrical gas-filled detector with an instrumented region featuring an inner radius of \sim 32\text{\,}\mathrm{cm} , outer radius of \sim 78\text{\,}\mathrm{cm} , and length of \sim 211\text{\,}\mathrm{cm} in a 1.4\text{\,}\mathrm{T} magnetic field. The TPC records the ionization signals produced as charged particles pass through the gas volume. Figure 2 illustrates how the ionized electrons drift under a uniform electric field produced by the central membrane toward the readout planes, where their arrival times encode the longitudinal ( z ) coordinate and the channel locations in the readout planes provide the transverse ( r – \phi ) coordinates. In this way, each charged particle leaves a series of high-resolution spacepoints ( \sim 150\text{\,}\mathrm{\SIUnitSymbolMicro m} in r – \phi ; \sim 750\text{\,}\mathrm{\SIUnitSymbolMicro m} in z ) along its trajectory [ 4 ] . As the central component of tracking in sPHENIX , the TPC provides precise momentum measurements from particle trajectories and particle identification. This dataset was compiled to provide a public, high-fidelity resource of simulated minimum-bias p{+}p collisions at \sqrt{s}=$200\text{\,}\mathrm{GeV}$ , motivated by the need for openly accessible data to develop and benchmark machine learning and reconstruction algorithms for TPC tracking. Such collisions, standard at RHIC, offer a controlled environment for testing methods before tackling complex heavy-ion collision data. This dataset ensures portability and reproducibility with clearly defined units and truth associations, fostering method development across physics and data science communities without requiring proprietary software access. The large unlabeled set ( 10\text{\,}\mathrm{M} events) is particularly well suited for training foundation models via self-supervised learning, addressing the need for scalable, high-fidelity datasets in charged-particle tracking. Figure 2: TPC schematic and collision event. Panel (a): The TPC schematic, where proton beams travel along the z direction. The electric field E drives ionization electrons to the readout planes, while the magnetic field B bends charged particle trajectories. Panel (b): An event display of reconstructed spacepoints in the TPC. Data Description The dataset repository contains simulated p{+}p collision data at \sqrt{s}=$200\text{\,}\mathrm{GeV}$ for the sPHENIX TPC, organized into three top-level folders. All data files are stored in .npz format, which can be read with Python libraries, such as NumPy . The TPC is a cylindrical detector with its z -axis along the beam direction, centered at the detector origin. Spacepoint coordinates (x,y,z) are reported in Cartesian coordinates (cm). Figure 3 provides a summary of the folder structure. 1. unlabeled/ : Contains 10\text{\,}\mathrm{M} minimum-bias events split into 100 shards (named spacepoints_[000-099].npz ), each with 100,000 events. Each file stores: (a) spacepoints : Array of shape (N_{i},4) with columns (E,x,y,z) , where x,y,z are positions ( \mathrm{cm} ) in the TPC active volume and E is the ionization signal (ADC units). 2. labeled/ : Contains 70\text{\,}\mathrm{k} events with full truth information, organized into three subsets: (a) train/ : 7 shards ( * _000.npz to * _006.npz ), each with 10\text{\,}\mathrm{k} events. (b) validation/ : A single file per array type for 13\text{\,}\mathrm{k} events. (c) test/ : A single file per array type for 7\text{\,}\mathrm{k} events. Each labeled subset includes four aligned arrays: (a) spacepoints : Array (N_{i},4) with (E,x,y,z) . (b) track_ids : Integer array (N_{i}) that assigns each spacepoint to a unique truth track identifier. (c) noise_tags : Integer array (N_{i}) that is equal to