Lexical Semantic Change through Large Language Models: a Survey

What is it about?

Recent advancements in Natural Language Processing (NLP) have highlighted the power of Large Language Models (LLMs) in analyzing how words change meaning over time, known as lexical semantic change. For instance, "gay" evolved from cheerful to homosexual. LLMs use word embeddings to capture these shifts by analyzing word contexts. Our survey focuses on how LLMs detect and model these changes, offering insights into linguistic evolution and societal reflection through automated, scalable methods.

Featured Image

Photo by Steve Johnson on Unsplash

Photo by Steve Johnson on Unsplash

Why is it important?

Our work stands out by focusing on the cutting-edge application of Large Language Models (LLMs) to automate the detection and modeling of lexical semantic change (LSC) in diachronic text corpora. Unlike traditional methods that rely on manual analysis, our approach harnesses the computational power of LLMs to process vast amounts of text efficiently. This enables us to uncover subtle shifts in word meanings over time, which is crucial for understanding how language evolves in response to societal changes. By presenting a comprehensive survey of LLM-based techniques for LSC detection, we provide a timely resource that not only advances computational linguistics but also appeals to a broader audience interested in the intersection of AI and historical language analysis.