Leveraging Multi-lingual Positive Instances in Contrastive Learning to Improve Sentence Embedding
This work addresses the challenge of multi-lingual sentence embedding for natural language processing applications, representing an incremental improvement over existing contrastive learning methods.
The paper tackled the problem of learning multi-lingual sentence embeddings by proposing MPCL, a method that leverages multiple positive instances in contrastive learning, resulting in improved retrieval, semantic similarity, and classification performances across various backbone models and downstream tasks, with better cross-lingual transfer in unseen languages.
Learning multi-lingual sentence embeddings is a fundamental task in natural language processing. Recent trends in learning both mono-lingual and multi-lingual sentence embeddings are mainly based on contrastive learning (CL) among an anchor, one positive, and multiple negative instances. In this work, we argue that leveraging multiple positives should be considered for multi-lingual sentence embeddings because (1) positives in a diverse set of languages can benefit cross-lingual learning, and (2) transitive similarity across multiple positives can provide reliable structural information for learning. In order to investigate the impact of multiple positives in CL, we propose a novel approach, named MPCL, to effectively utilize multiple positive instances to improve the learning of multi-lingual sentence embeddings. Experimental results on various backbone models and downstream tasks demonstrate that MPCL leads to better retrieval, semantic similarity, and classification performances compared to conventional CL. We also observe that in unseen languages, sentence embedding models trained on multiple positives show better cross-lingual transfer performance than models trained on a single positive instance.