Can an LLM speculate on name origins using the same kind of "when and where" data a human expert might use? Here's an in-depth writeup of my attempt to find out, including all the prompts that went into the two-stage workflow I designed:
https://nameplay.org/blog/educating-name-meaning-guesses-with-data
And here's an interactive directory with links to the inferred origins, for your reviewing entertainment: https://nameplay.org/list/names-with-inferred-origins
I'm curious to hear whether you think this attempt to produce less-sloppy content using an LLM was successful, or whether I've just added to the mountain of name-related slop already on the internet...?
[link] [comments]