ReFRAME or Remain: Unsupervised Lexical Semantic Change Detection with Frame Semantics
The episode discusses the paper 'ReFRAME or Remain,' which focuses on unsupervised lexical semantic change detection using Frame Semantics. The authors propose a method to track how language evolves by quantifying divergence from established usage patterns in historical text data. This approach allows researchers to pinpoint exactly how and why specific word meanings shift over time, moving beyond speculation into measurable science.