Tag: Low-Rank Adaptation

All the papers with the tag "Low-Rank Adaptation".

• HSplitLoRA: A Heterogeneous Split Parameter-Efficient Fine-Tuning Framework for Large Language Models

  grok-3-latest

  Score: 0.78

  Published:2025年5月5日 at  at 17:09

  #LLM, #Parameter Efficient Tuning, #Split Learning, #Heterogeneous Computing, #Low-Rank Adaptation

  HSplitLoRA提出了一种异构参数高效微调框架，通过分割学习和LoRA结合，动态调整重要权重、适配器秩和模型分割点，并在异构环境下显著提升大型语言模型的训练精度和收敛速度。