Unsupervised Attention-based Sentence-Level Meta-Embeddings from Contextualised Language Models
This work addresses the need for robust sentence representations in NLP, offering a task-agnostic approach that is incremental in enhancing existing meta-embedding techniques.
The paper tackles the problem of varying performance in downstream NLP applications by different contextualized language models, proposing an unsupervised sentence-level meta-embedding method that combines multiple models to preserve their complementary strengths, resulting in improved performance on semantic textual similarity benchmarks compared to previous methods and a supervised baseline.
A variety of contextualised language models have been proposed in the NLP community, which are trained on diverse corpora to produce numerous Neural Language Models (NLMs). However, different NLMs have reported different levels of performances in downstream NLP applications when used as text representations. We propose a sentence-level meta-embedding learning method that takes independently trained contextualised word embedding models and learns a sentence embedding that preserves the complementary strengths of the input source NLMs. Our proposed method is unsupervised and is not tied to a particular downstream task, which makes the learnt meta-embeddings in principle applicable to different tasks that require sentence representations. Specifically, we first project the token-level embeddings obtained by the individual NLMs and learn attention weights that indicate the contributions of source embeddings towards their token-level meta-embeddings. Next, we apply mean and max pooling to produce sentence-level meta-embeddings from token-level meta-embeddings. Experimental results on semantic textual similarity benchmarks show that our proposed unsupervised sentence-level meta-embedding method outperforms previously proposed sentence-level meta-embedding methods as well as a supervised baseline.