PulseAugur
实时 10:51:50
English(EN) Pre-Training for Simulation-Based Science: A Study on Jet Foundation Model Training Objectives

新研究比较高能物理领域人工智能的预训练方法

一篇新的arXiv论文探讨了在基于仿真的科学(特别是高能物理领域)中基础模型的预训练目标。该研究使用OmniLearned High Energy Physics FM框架,比较了监督分类、流匹配生成和自监督掩码粒子建模。结果表明,当标签丰富时,纯分类器预训练效果最佳,但在标签稀疏的情况下,将其与掩码粒子建模相结合非常有效。对于生成任务,必须在预训练中包含流匹配才能在下游任务中获得显著优势。 AI

排序理由 该集群包含一篇详细介绍人工智能模型训练目标研究结果的学术论文。[lever_c_demoted from research: ic=1 ai=1.0]

在 arXiv cs.LG 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.LG TIER_1 English(EN) · Ibrahim Elsharkawy, Joschka Birk, Vinicius Mikuni, Wahid Bhimji, Gregor Kasieczka, Benjamin Nachman ·

    Pre-Training for Simulation-Based Science: A Study on Jet Foundation Model Training Objectives

    arXiv:2606.14870v1 Announce Type: cross Abstract: Foundation models (FMs) trained on large datasets and fine-tuned on downstream tasks have emerged as a powerful paradigm in AI for science. Industrial FMs are typically trained using self-supervision with masking due to the lack o…