PCL: Peer-Contrastive Learning with Diverse Augmentations for Unsupervised Sentence Embeddings
This work addresses a key bottleneck in unsupervised NLP for tasks like semantic similarity, though it is incremental as it builds on existing contrastive learning methods.
The paper tackles the problem of learning high-quality unsupervised sentence embeddings by addressing the bias introduced by mono-augmentation strategies in contrastive learning, proposing Peer-Contrastive Learning (PCL) with diverse augmentations, which achieves improved performance on STS benchmarks.
Learning sentence embeddings in an unsupervised manner is fundamental in natural language processing. Recent common practice is to couple pre-trained language models with unsupervised contrastive learning, whose success relies on augmenting a sentence with a semantically-close positive instance to construct contrastive pairs. Nonetheless, existing approaches usually depend on a mono-augmenting strategy, which causes learning shortcuts towards the augmenting biases and thus corrupts the quality of sentence embeddings. A straightforward solution is resorting to more diverse positives from a multi-augmenting strategy, while an open question remains about how to unsupervisedly learn from the diverse positives but with uneven augmenting qualities in the text field. As one answer, we propose a novel Peer-Contrastive Learning (PCL) with diverse augmentations. PCL constructs diverse contrastive positives and negatives at the group level for unsupervised sentence embeddings. PCL performs peer-positive contrast as well as peer-network cooperation, which offers an inherent anti-bias ability and an effective way to learn from diverse augmentations. Experiments on STS benchmarks verify the effectiveness of PCL against its competitors in unsupervised sentence embeddings.