PulseAugur
实时 13:13:55
English(EN) JetParticle-JEPA: An Efficient Self-Supervised Representation Learning method for Jet Tagging in High-Energy Physics

新的自监督方法增强了高能物理中的喷注辨识能力

研究人员开发了 JetParticle-JEPA (JP-JEPA),一种用于高能物理喷注辨识的新型自监督学习方法。该方法基于 Particle Transformer 构建,可以直接从粒子数据中学习有意义的表征,而无需广泛的标记数据集。在 JetClass 等基准测试中,JP-JEPA 的性能与监督方法相当,并且在面对探测器错误建模和数据限制时表现出更强的鲁棒性。 AI

排序理由 该聚类包含一篇学术论文,详细介绍了特定科学领域的新方法。[lever_c_demoted from research: ic=1 ai=0.7]

在 arXiv cs.AI 阅读 →

AI 生成摘要 · Google Gemini · 来自 1 个来源。 我们如何撰写摘要 →

报道来源 [1]

  1. arXiv cs.AI TIER_1 English(EN) · Guillaume Letellier (LPCC), Antonin Vacheret (LPCC), Fr\'ed\'eric Jurie ·

    JetParticle-JEPA: An Efficient Self-Supervised Representation Learning method for Jet Tagging in High-Energy Physics

    arXiv:2606.14813v1 Announce Type: cross Abstract: Jet tagging at the Large Hadron Collider increasingly relies on deep learning models trained on massive simulated datasets, leading to high computational costs and limited robustness to detector mismodeling. We introduce JetPartic…