Recent advances have led to the availability of many pre-trained language models (PLMs); however, a question that remains is how much data is truly needed to fine-tune PLMs for downstream tasks? In this work, we introduce DEFT, a data-efficient fine-tuning framework that leverages unsupervised core-set selection to minimize the amount of data needed to fine-tune PLMs for downstream tasks. We demonstrate the efficacy of our DEFT framework in the context of text-editing LMs, and compare to the state-of-the art text-editing model, CoEDIT (Raheja et al., 2023). Our quantitative and qualitative results demonstrate that DEFT models are just as accurate as CoEDIT while being finetuned on ~70% less data.

翻译：摘要：近期研究进展使得大量预训练语言模型（PLMs）得以可用；然而，一个尚待解决的问题是：微调PLMs以适配下游任务究竟需要多少数据？本文提出DEFT——一种数据高效微调框架，通过无监督核心集选择最小化微调PLMs所需的数据量。我们以文本编辑语言模型为场景验证DEFT框架的有效性，并与当前最先进的文本编辑模型CoEDIT（Raheja等，2023）进行对比。定性与定量结果表明，DEFT模型在微调数据量减少约70%的情况下，仍能达到与CoEDIT同等的精度。