A Survey on Word Meta-Embedding Learning
It provides a foundational resource for NLP practitioners by organizing and analyzing existing meta-embedding techniques, though it is incremental as a survey without new experimental results.
This paper addresses the lack of a systematic survey on meta-embedding learning, which combines multiple source word embeddings to improve accuracy, by classifying methods based on factors like static vs. contextualized embeddings and unsupervised vs. task-specific training.
Meta-embedding (ME) learning is an emerging approach that attempts to learn more accurate word embeddings given existing (source) word embeddings as the sole input. Due to their ability to incorporate semantics from multiple source embeddings in a compact manner with superior performance, ME learning has gained popularity among practitioners in NLP. To the best of our knowledge, there exist no prior systematic survey on ME learning and this paper attempts to fill this need. We classify ME learning methods according to multiple factors such as whether they (a) operate on static or contextualised embeddings, (b) trained in an unsupervised manner or (c) fine-tuned for a particular task/domain. Moreover, we discuss the limitations of existing ME learning methods and highlight potential future research directions.