Semantic Specialization for Knowledge-based Word Sense Disambiguation
This work addresses the challenge of disambiguating word senses in natural language processing, which is crucial for improving language understanding in AI systems, and it represents an incremental advancement by building on existing embedding adaptation approaches.
The paper tackled the problem of knowledge-based Word Sense Disambiguation (WSD) by proposing a semantic specialization method that adapts contextualized embeddings using lexical knowledge, achieving state-of-the-art performance on knowledge-based WSD when combined with a reranking heuristic.
A promising approach for knowledge-based Word Sense Disambiguation (WSD) is to select the sense whose contextualized embeddings computed for its definition sentence are closest to those computed for a target word in a given sentence. This approach relies on the similarity of the \textit{sense} and \textit{context} embeddings computed by a pre-trained language model. We propose a semantic specialization for WSD where contextualized embeddings are adapted to the WSD task using solely lexical knowledge. The key idea is, for a given sense, to bring semantically related senses and contexts closer and send different/unrelated senses farther away. We realize this idea as the joint optimization of the Attract-Repel objective for sense pairs and the self-training objective for context-sense pairs while controlling deviations from the original embeddings. The proposed method outperformed previous studies that adapt contextualized embeddings. It achieved state-of-the-art performance on knowledge-based WSD when combined with the reranking heuristic that uses the sense inventory. We found that the similarity characteristics of specialized embeddings conform to the key idea. We also found that the (dis)similarity of embeddings between the related/different/unrelated senses correlates well with the performance of WSD.