25 April 2025 in The European Physical Journal C
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.
7 October 2025 in EPJ Web of Conferences
Track reconstruction is a crucial part of High Energy Physics experiments. Traditional methods for the task, relying on Kalman Filters, scale poorly with detector occupancy. In the context of the upcoming High Luminosity-LHC, solutions based on Machine Learning (ML) and deep learning are very appealing. We investigate the feasibility of training multiple ML architectures to infer track-defining parameters from detector signals, for the application of offline reconstruction. We study and compare three Transformer model designs, as well as a U-Net architecture. We describe in detail the two most promising approaches and benchmark the pipelines for physics performance and inference speed on methodically simplified datasets, generated by the recently developed simulation framework, REDuced VIrtual Detector (REDVID). Our second batch of simplified datasets are derived from the TrackML dataset. Our preliminary results show promise for the application of such deep learning techniques on more realistic data for tracking, as well as efficient elimination of solutions.