Random Masking Finds Winning Tickets for Parameter Efficient Fine-tuning

arXiv:2405.02596v1 Announce Type: new
Abstract: Fine-tuning large language models (LLM) can be costly. Parameter-efficient fine-tuning (PEFT) addresses the problems by training a fraction of the parameters, whose success reveals the expressiveness and flexibility of pretrained models. This paper studies the limit of PEFT, by further simplifying its design and reducing the number of trainable parameters beyond standard setups. To this end, we use Random Masking to fine-tune the pretrained model. Despite its simplicity, we show that Random Masking is surprisingly …

abstract arxiv cs.ai cs.cl cs.lg design fine-tuning flexibility language language models large language large language models llm masking paper parameters peft pretrained models random simplifying studies success tickets training type