laserduck

I agree that it makes sense to group the Indic languages together due to cultural proximity but why would you group the Indic languages with Middle Eastern ones? Might as well group it with European or African or Sinitic languages at that point.

I wonder why they grouped languages from the Middle East and South Asia together. Arabic and Hebrew are Semitic languages - no language from that family tree is native to the subcontinent. It would make sense if northern languages like Hindi, Urdu, Bengali, Nepali, etc were grouped with Persian, French, Russian, etc since those are all from the Indo-European family. South Indian languages like Telugu and Tamil are from a completely different family (Dravidian).

Why not either train the model exclusively on Semitic languages for further performance for those languages or on a wider set of languages for better multilingual performance overall? I don't understand the logic here.

Thank you so much! Do you know they would be willing to work with people who are still early in the prototyping process? Will they only help with PCB design or can they also help with casing, etc?

Really appreciate your response!

I love this channel! The videos are well made narratively while still preseving the facts and citing sources. Almost the ideal combo of being academic and entertaining