Fine-tuning Happens in Tiny Subspaces: Exploring Intrinsic Task-specific Subspaces of Pre-trained Language Models
This addresses the issue of parameter redundancy in PLMs for NLP practitioners, offering a method to reduce fine-tuning complexity, though it is incremental in optimizing existing fine-tuning approaches.
The paper tackles the problem of fine-tuning pre-trained language models by discovering intrinsic task-specific subspaces, showing that models can be effectively fine-tuned with a small number of free parameters and identifying outlier dimensions crucial for task performance.
Pre-trained language models (PLMs) are known to be overly parameterized and have significant redundancy, indicating a small degree of freedom of the PLMs. Motivated by the observation, in this paper, we study the problem of re-parameterizing and fine-tuning PLMs from a new perspective: Discovery of intrinsic task-specific subspace. Specifically, by exploiting the dynamics of the fine-tuning process for a given task, the parameter optimization trajectory is learned to uncover its intrinsic task-specific subspace. A key finding is that PLMs can be effectively fine-tuned in the subspace with a small number of free parameters. Beyond, we observe some outlier dimensions emerging during fine-tuning in the subspace. Disabling these dimensions degrades the model performance significantly. This suggests that these dimensions are crucial to induce task-specific knowledge to downstream tasks.