Saved in:
| Main Authors: | , , , , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2026
|
| Subjects: | |
| Online Access: | https://arxiv.org/abs/2602.00141 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866911412301856768 |
|---|---|
| author | Kim, Daeun Lee, Jiwon Jeong, Wonjun Noh, Hyeongwoo Kim, Giyeong Cho, Jaeyoon Kwak, Geonhee Yang, Seunghwan Kweon, MinJung |
| author_facet | Kim, Daeun Lee, Jiwon Jeong, Wonjun Noh, Hyeongwoo Kim, Giyeong Cho, Jaeyoon Kwak, Geonhee Yang, Seunghwan Kweon, MinJung |
| contents | We present a comprehensive comparison of convolutional and transformer-based models for distinguishing quark and gluon jets using simulated jet images from Pythia 8. By encoding jet substructure into a three-channel representation of particle kinematics, we evaluate the performance of convolutional neural networks (CNNs), Vision Transformers (ViTs), and Swin Transformers (Swin-Tiny) under both supervised and self-supervised learning setups. Our results show that fine-tuning only the final two transformer blocks of the Swin-Tiny model achieves the best trade-off between efficiency and accuracy, reaching 81.4% accuracy and an AUC (area under the ROC curve) of 88.9%. Self-supervised pretraining with Momentum Contrast (MoCo) further enhances feature robustness and reduces the number of trainable parameters. These findings highlight the potential of hierarchical attention-based models for jet substructure studies and for domain transfer to real collision data. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2602_00141 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | Comparison of Image Processing Models in Quark Gluon Jet Classification Kim, Daeun Lee, Jiwon Jeong, Wonjun Noh, Hyeongwoo Kim, Giyeong Cho, Jaeyoon Kwak, Geonhee Yang, Seunghwan Kweon, MinJung Data Analysis, Statistics and Probability Computer Vision and Pattern Recognition Machine Learning High Energy Physics - Experiment We present a comprehensive comparison of convolutional and transformer-based models for distinguishing quark and gluon jets using simulated jet images from Pythia 8. By encoding jet substructure into a three-channel representation of particle kinematics, we evaluate the performance of convolutional neural networks (CNNs), Vision Transformers (ViTs), and Swin Transformers (Swin-Tiny) under both supervised and self-supervised learning setups. Our results show that fine-tuning only the final two transformer blocks of the Swin-Tiny model achieves the best trade-off between efficiency and accuracy, reaching 81.4% accuracy and an AUC (area under the ROC curve) of 88.9%. Self-supervised pretraining with Momentum Contrast (MoCo) further enhances feature robustness and reduces the number of trainable parameters. These findings highlight the potential of hierarchical attention-based models for jet substructure studies and for domain transfer to real collision data. |
| title | Comparison of Image Processing Models in Quark Gluon Jet Classification |
| topic | Data Analysis, Statistics and Probability Computer Vision and Pattern Recognition Machine Learning High Energy Physics - Experiment |
| url | https://arxiv.org/abs/2602.00141 |