Quantum Computing and Machine Learning Workshop 2025

A foundation jet model aims to achieve optimal performance across all jet analysis tasks while ensuring strong generalization. Building onSophon, a pre-trained jet classification model, we developSophon++, which employs contrastive learning to connect initial, parton-level, and reconstruction-level particles, enabling continuous encoding of generator-level particle configurations into the model’s latent space. While matchingSophonin classification performance,Sophon++demonstrates stronger generalization through several fine-tuning tasks. This work provides a promising pathway towards a foundation jet model for analysis.