reCSE: Portable Reshaping Features for Sentence Embedding in Self-supervised Contrastive Learning
This work addresses efficiency and representation issues in sentence embedding models, offering a portable method that can enhance other frameworks, though it appears incremental as it builds on existing contrastive learning approaches.
The authors tackled the problem of representation polarity and GPU memory consumption in self-supervised contrastive learning for sentence embeddings by proposing reCSE, a framework that reshapes input features to aggregate global token information, achieving competitive performance in semantic similarity tasks.
We propose reCSE, a self supervised contrastive learning sentence representation framework based on feature reshaping. This framework is different from the current advanced models that use discrete data augmentation methods, but instead reshapes the input features of the original sentence, aggregates the global information of each token in the sentence, and alleviates the common problems of representation polarity and GPU memory consumption linear increase in current advanced models. In addition, our reCSE has achieved competitive performance in semantic similarity tasks. And the experiment proves that our proposed feature reshaping method has strong universality, which can be transplanted to other self supervised contrastive learning frameworks and enhance their representation ability, even achieving state-of-the-art performance. Our code is available at https://github.com/heavenhellchen/reCSE.