WiC: the Word-in-Context Dataset for Evaluating Context-Sensitive Meaning Representations
This provides a new benchmark for evaluating context-sensitive representations in NLP, addressing a gap in the field, though it is incremental as it builds on prior work.
The authors tackled the lack of evaluation benchmarks for context-sensitive word meaning representations by creating WiC, a large-scale dataset based on expert annotations, which addresses the shortcomings of existing datasets like Stanford Contextual Word Similarity.
By design, word embeddings are unable to model the dynamic nature of words' semantics, i.e., the property of words to correspond to potentially different meanings. To address this limitation, dozens of specialized meaning representation techniques such as sense or contextualized embeddings have been proposed. However, despite the popularity of research on this topic, very few evaluation benchmarks exist that specifically focus on the dynamic semantics of words. In this paper we show that existing models have surpassed the performance ceiling of the standard evaluation dataset for the purpose, i.e., Stanford Contextual Word Similarity, and highlight its shortcomings. To address the lack of a suitable benchmark, we put forward a large-scale Word in Context dataset, called WiC, based on annotations curated by experts, for generic evaluation of context-sensitive representations. WiC is released in https://pilehvar.github.io/wic/.