Measuring Word Meaning Change Across Time and Speaker Age

Gaurav Kamath

Mila and McGill

The NLP Reading Group is happy to receive Gaurav Kamath who will be giving a talk in-person at Mila on “Measuring Word Meaning Change Across Time and Speaker Age”.

Recording

Talk Description

A central question in the study of language change is whether or not it is generational. Under this picture, language change is an iterative, generation-by-generation process: new generations of speakers introduce innovations, while older generations maintain their prior linguistic patterns, and the language evolves as the newer generations replace older ones. Conversely, language change could be a zeitgeist phenomenon, in which changes are universally adopted by speakers simultaneously, across ages and generational groups. In this talk, I present work recently published in PNAS, which asks this question in the context of word meaning change. We analyze meaning change in over 100 words across more than 7.9 million U.S. congressional speeches, to observe whether, when a word sense rises or falls in prominence, adult speakers from different generations uniformly adopt it, or those from older generations conserve their prior usage. We use masked language models to identify different senses of each word, and then model the prevalence of each of these word senses as a function of time and speaker age. We find that most words show only a small effect of speaker age; across almost 140 years of Congress, older speakers typically take longer than younger speakers to follow changes in word usage, but nevertheless do so within a few years. Our findings suggest that despite minor age-based differences, word meaning change among adults is broadly a zeitgeist process, and that older adult speakers are able readily able to adopt new word usage patterns.

Speaker Bio

Gaurav Kamath is a 4th-year PhD student in Linguistics at McGill University and Mila. His research interests are at the intersection of NLP and linguistics, including work that compares the linguistic behaviour of LLMs and humans, as well as attempts to use language models to answer questions about human language.

Logistics

Date: September 26th
Time: 2:00PM
Location: H04 or via Google Meet (See email)