Improving language models fine-tuning with representation consistency targets. (arXiv:2205.11603v1 [cs.CL])

Fine-tuning contextualized representations learned by pre-trained language
models has become a standard practice in the NLP field. However, pre-trained
representations are prone to degradation (also known as representation
collapse) during fine-tuning, which leads to instability, suboptimal
performance, and weak generalization. In this paper, we propose a novel
fine-tuning method that avoids representation collapse during fine-tuning by
discouraging undesirable changes in the representations. We show that our
approach matches or exceeds the performance of the existing
regularization-based fine-tuning methods across 13 …

arxiv fine-tuning language language models representation