%0 Report
%T Tracking in Dense Environments with Transformers
%N ATL-PHYS-PUB-2025-045
%M PUBDB-2025-04910
%M ATL-PHYS-PUB-2025-045
%P 12
%D 2025
%X This work presents a novel application of machine learning to the pattern-recognition stage of charged-particle reconstruction, enabling learned hit-to-track association within dense environments, such as the cores of high-pTjets. Our Transformer-based architecture is based on a MaskFormer model that jointly optimises hit assignments and the estimation of the charged particles' properties. Trained and evaluated in dense environments the model delivers up to a 30
%F PUB:(DE-HGF)29
%9 Report
%R 10.3204/PUBDB-2025-04910
%U https://bib-pubdb1.desy.de/record/640807