Deriving Word Vectors from Contextualized Language Models using Topic-Aware Mention Selection
This work addresses the problem of improving word representation quality for natural language processing tasks, offering an incremental advancement over prior methods.
The paper tackles the challenge of learning word representations that reflect semantic properties by proposing a method that uses contextualized language models and topic-aware mention selection to derive topic-specific word vectors, resulting in vectors that are more predictive of semantic properties than existing word embeddings and CLM-based strategies.
One of the long-standing challenges in lexical semantics consists in learning representations of words which reflect their semantic properties. The remarkable success of word embeddings for this purpose suggests that high-quality representations can be obtained by summarizing the sentence contexts of word mentions. In this paper, we propose a method for learning word representations that follows this basic strategy, but differs from standard word embeddings in two important ways. First, we take advantage of contextualized language models (CLMs) rather than bags of word vectors to encode contexts. Second, rather than learning a word vector directly, we use a topic model to partition the contexts in which words appear, and then learn different topic-specific vectors for each word. Finally, we use a task-specific supervision signal to make a soft selection of the resulting vectors. We show that this simple strategy leads to high-quality word vectors, which are more predictive of semantic properties than word embeddings and existing CLM-based strategies.