让LoRA具备非线性表达能力，提升微调效果同时保持可合并性。
AFA-LoRA: Enabling Non-Linear Adaptations in LoRA with Activation Function Annealing
- 通过渐变激活函数，让LoRA训练初期用非线性、后期转为线性。
- 在监督微调、强化学习等任务中，性能接近全参数微调。
- 适合需要高效微调又追求高表现的模型应用者。
低秩适配（LoRA）是一种广泛应用的参数高效微调（PEFT）方法，但其线性适配过程限制了表达能力。本文提出AFA-LoRA，一种新训练策略，使LoRA在保持无缝合并性的同时引入非线性表达能力。核心创新是采用渐变激活函数，在训练过程中从非线性逐渐过渡到线性，使适配器初始阶段具备更强表征能力，最终收敛为可合并的线性形式。我们在监督微调、强化学习和推测解码任务上实现了该方法。结果表明，AFA-LoRA显著缩小了LoRA与全参数训练之间的性能差距，推动了一种更强大且实用的参数高效适配范式。
原文摘要 · Abstract (English)
Low-Rank Adaptation (LoRA) is a widely adopted parameter-efficient fine-tuning (PEFT) method. However, its linear adaptation process limits its expressive power. This means there is a gap between the expressive power of linear training and non-linear training. To bridge this gap, we propose AFA-LoRA, a novel training strategy that brings non-linear expressivity to LoRA while maintaining its seamless mergeability. Our key innovation is an annealed activation function that transitions from a non-linear to a linear transformation during training, allowing the adapter to initially adopt stronger representational capabilities before converging to a mergeable linear form. We implement our method on supervised fine-tuning, reinforcement learning, and speculative decoding. The results show that AFA-LoRA reduces the performance gap between LoRA and full-parameter training. This work enables a more powerful and practical paradigm of parameter-efficient adaptation.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。