Representation Projection Invariance Mitigates Representation Collapse

Anastasia Razdaibiedina, Ashish Khetan, Zohar Karnin, Daniel Khashabi, Vivek Madan

Submission Type: Regular Long Paper

Submission Track: Machine Learning for NLP

Keywords: representation learning, generalization, representation collapse

TL;DR: We propose a new regularization method that prevents collapse of representations and improves generalization of the model.

Abstract: Fine-tuning contextualized representations learned by pre-trained language models remains a prevalent practice in NLP. However, fine-tuning can lead to representation degradation (also known as representation collapse), which may result in instability, sub-optimal performance, and weak generalization. In this paper, we propose Representation Projection Invariance (REPINA), a novel regularization method to maintain the information content of representation and reduce representation collapse during fine-tuning by discouraging undesirable changes in the representations. We study the empirical behavior of the proposed regularization in comparison to 5 comparable baselines across 13 language understanding tasks (GLUE benchmark and six additional datasets). When evaluating in-domain performance, REPINA consistently outperforms other baselines on most tasks (10 out of 13). Additionally, REPINA improves out-of-distribution performance. We also demonstrate its effectiveness in few-shot settings and robustness to label perturbation. As a by-product, we extend previous studies of representation collapse and propose several metrics to quantify it. Our empirical findings show that our approach is significantly more effective at mitigating representation collapse.

Submission Number: 4706