Evaluating Unsupervised Dimensionality Reduction Methods for Pretrained Sentence Embeddings
This addresses memory and compute constraints for deploying sentence embeddings in resource-limited devices, but it is incremental as it applies existing methods to a new context.
The paper tackles the problem of high dimensionality in pretrained language model sentence embeddings by evaluating unsupervised dimensionality reduction methods, finding that PCA can reduce dimensionality by almost 50% without significant performance loss and sometimes improves performance.
Sentence embeddings produced by Pretrained Language Models (PLMs) have received wide attention from the NLP community due to their superior performance when representing texts in numerous downstream applications. However, the high dimensionality of the sentence embeddings produced by PLMs is problematic when representing large numbers of sentences in memory- or compute-constrained devices. As a solution, we evaluate unsupervised dimensionality reduction methods to reduce the dimensionality of sentence embeddings produced by PLMs. Our experimental results show that simple methods such as Principal Component Analysis (PCA) can reduce the dimensionality of sentence embeddings by almost $50\%$, without incurring a significant loss in performance in multiple downstream tasks. Surprisingly, reducing the dimensionality further improves performance over the original high-dimensional versions for the sentence embeddings produced by some PLMs in some tasks.