Abstract:High-Energy Physics experiments are facing a multi-fold data increase with every new iteration. This is certainly the case for the upcoming High-Luminosity LHC upgrade. Such increased data processing requirements forces revisions to almost every step of the data processing pipeline. One such step in need of an overhaul is the task of particle track reconstruction, a.k.a., tracking. A Machine Learning-assisted solution is expected to provide significant improvements, since the most time-consuming step in tracking is the assignment of hits to particles or track candidates. This is the topic of this paper. We take inspiration from large language models. As such, we consider two approaches: the prediction of the next word in a sentence (next hit point in a track), as well as the one-shot prediction of all hits within an event. In an extensive design effort, we have experimented with three models based on the Transformer architecture and one model based on the U-Net architecture, performing track association predictions for collision event hit points. In our evaluation, we consider a spectrum of simple to complex representations of the problem, eliminating designs with lower metrics early on. We report extensive results, covering both prediction accuracy (score) and computational performance. We have made use of the REDVID simulation framework, as well as reductions applied to the TrackML data set, to compose five data sets from simple to complex, for our experiments. The results highlight distinct advantages among different designs in terms of prediction accuracy and computational performance, demonstrating the efficiency of our methodology. Most importantly, the results show the viability of a one-shot encoder-classifier based Transformer solution as a practical approach for the task of tracking.

Training toward significance with the decorrelated event classifier transformer neural network

Training toward significance with the decorrelated event classifier transformer neural network

Attention to the strengths of physical interactions: Transformer and graph-based event classification for particle physics experiments

Transformer networks for Heavy flavor jet tagging

Interpreting Transformers for Jet Tagging

Multi-scale cross-attention transformer encoder for event classification

Learning to see R-parity violating scalar top decays

TrackFormers: In Search of Transformer-Based Particle Tracking for the High-Luminosity LHC Era

Interpretable correlator Transformer for image-like quantum matter data

Interpretable Joint Event-Particle Reconstruction for Neutrino Physics at NOvA with Sparse CNNs and Transformers

Learning the language of QCD jets with transformers

Pretrained Event Classification Model for High Energy Physics Analysis

Transformers for Charged Particle Track Reconstruction in High Energy Physics

Hybrid Quantum Vision Transformers for Event Classification in High Energy Physics

Invariant Representation Driven Neural Classifier for Anti-QCD Jet Tagging

Learning stochastic dynamics and predicting emergent behavior using transformers

From strange-quark tagging to fragmentation tagging with machine learning

Improved particle-flow event reconstruction with scalable neural networks for current and future particle detectors

Learning Efficient Representations of Neutrino Telescope Events