研究人员开发了MERIT，一种新颖的去中心化指令调优管道，旨在克服大型语言模型中的梯度干扰和同步瓶颈。该方法包括估计数据集级别的梯度冲突，沿着这些冲突轴划分模型混合体，然后独立微调每个分区，最后进行一次合并步骤。MERIT在多模态和纯文本任务上展示了改进的性能，其效果与集中式训练相当或更优，同时通信开销显著降低。 AI
影响 MERIT等去中心化训练方法可以显著降低微调大型模型相关的计算和通信成本。
排序理由 该集群包含一篇详细介绍LLM指令调优新方法的学术论文。
在 Hugging Face Daily Papers 阅读 →
AI 生成摘要 · Google Gemini · 来自 2 个来源。 我们如何撰写摘要 →
arXiv:2606.01717v1 Announce Type: new Abstract: Instruction tuning aligns large language models, including multimodal ones, with diverse user intents, but scaling to heterogeneous mixtures is hindered by gradient interference and bandwidth-heavy synchronization. We ask whether th…
Instruction tuning of large language models can be improved through decentralized training that partitions mixed datasets based on gradient conflicts and merges results via weighted averaging, achieving performance comparable to centralized methods with reduced communication over…