this post was submitted on 11 Jan 2025

229 points (97.9% liked)

Data is Beautiful

1346 readers

3 users here now

Be respectful

founded 6 months ago

MODERATORS

KnowledgeIsPower@mander.xyz

Beaver@lemmy.ca

WanderingPhoenix@mander.xyz

229

A representation of words derived from Latin (files.catbox.moe)

submitted 4 days ago* (last edited 3 days ago) by fxomt@lemm.ee to c/dataisbeautiful@mander.xyz

45 comments fedilink hide all child comments

Cross posted from: Latin@lemm.ee

lingua latina pater linguarum dimidum est 😎

I hope it's okay for me to crosspost here.

you are viewing a single comment's thread
view the rest of the comments

[–] Hackworth@lemmy.world 5 points 3 days ago (2 children)

I wonder if something like the semantic tokenization method would benefit from using etymological data like this, particularly for a multilingual llm.

[–] gandalf_der_12te@discuss.tchncs.de 3 points 3 days ago* (last edited 3 days ago)

i know that my NN internally uses semantic tokenization method.

i literally often seek the word roots when talking to somebody. it helps me focus.

[–] fxomt@lemm.ee 2 points 3 days ago

Interesting paper, thanks for sharing