A Step Toward Federated Pretraining of Multimodal Large Language Models
迈向多模态大语言模型联邦预训练的一小步
机构 * MAIS, Institute of Automation, Chinese Academy of Sciences(中国科学院自动化研究所多模态人工智能系统实验室) ; King Abdullah University of Science and Technology (KAUST)(阿卜杜拉国王科技大学) ; Pengcheng Laboratory(鹏城实验室) ; School of Artificial Intelligence, University of Chinese Academy of Sciences (UCAS)(中国科学院大学人工智能学院)
专题命中 预训练与数据 :large language model(title,abstract);language model(title,abstract);pretraining(title);LLM(abstract)
AI总结 本文提出Fed-MA任务,通过冻结视觉编码器和LLM,协同训练跨模态投影器,解决参数干扰和梯度震荡问题,提出Fed-CMP框架在联邦预训练中取得显著优势。