A Language Model for Particle Tracking
This work addresses the problem of task-specific models in particle physics by introducing a foundational model for detector understanding, representing an incremental step in the field.
The authors tackled the lack of generalization in deep learning models for particle tracking by proposing a unified language model approach, resulting in TrackingBERT, which offers latent embeddings for cross-task applications.
Particle tracking is crucial for almost all physics analysis programs at the Large Hadron Collider. Deep learning models are pervasively used in particle tracking related tasks. However, the current practice is to design and train one deep learning model for one task with supervised learning techniques. The trained models work well for tasks they are trained on but show no or little generalization capabilities. We propose to unify these models with a language model. In this paper, we present a tokenized detector representation that allows us to train a BERT model for particle tracking. The trained BERT model, namely TrackingBERT, offers latent detector module embedding that can be used for other tasks. This work represents the first step towards developing a foundational model for particle detector understanding.