tugdual

I love this ! Used to work at Rain AI on training neural networks in unconventional hardware - people often that computers don't necessarily have to be electronic digital - there is a whole domain dedicated to creating machines that can apply certain mathematical operations faster or more efficiently than their electronic counter parts. I created this site to try create a classification of that space:

https://computers.tugdual.fr/

Absolutely incredible

They’re not competing in the same domain - if you look at their business model it actually is much closer to ML consulting for companies (CMA CGM, ASML, Airbus…). The big three are trying to capture B2C mainly while Mistral is full focused B2B

Love it !

The reason we're conflating them is because there is a strong correlation between "highly processed food" and "designed recklessly". If you look at Carlos Monteiro (The pioneer in this domain) he operationalized it with the NOVA metric. NOVA 4 being the closest to what you're talking about:

"Industrially manufactured food products made up of several ingredients (formulations) including sugar, oils, fats and salt (generally in combination and in higher amounts than in processed foods) and food substances of no or rare culinary use (such as high-fructose corn syrup, hydrogenated oils, modified starches and protein isolates)..." [1]

[1] https://en.wikipedia.org/wiki/Nova_classification

I highly recommend Chris van Tulleken's Ultra-processed people for a more indepth read on this fat correlation (excuse the pun :))

Amazing work, I have a deep respect for this type of project where the underlying theme is perseverance !

This is what my Master project was about, working in the case of Wolof. I've trained XTTSv2 and had solid results with less than 20h of paired data that wasn't of the highest quality either - hmu: [email protected]

This is an interesting problem that has various challenges - currently most tokenization solutions where trainees using hype pair encoding where the most commonly seen combinations of letters were being selected to be a mapping. This meant that the majority of tokenization was English mappings meaning your LLM had a better tokenization of English compared to other languages it was being trained on.

C.f. https://medium.com/@biswanai92/understanding-token-fertility...

Got a demo ?

When Carlos Monteiro decided to operationalize UPFs by giving them a definition (laymans terms: UPF is one ingredient you wouldn't find in a traditional kitchen and wrapped in plastic) Kevin Hall from the US had the same reaction as you and decided to make a multi-million dollar experiment to disprove the definition proposed by Dr. Monteiro. Result: People who ate unprocessed lost weight, and the other group gained weight. (Groups were exchanged after 2 weeks and saw similar effects).

They could be solving it with multimodal mixup, a technique making sure that there's no big latent gap between the two : https://arxiv.org/abs/2203.03897

I actually did something similar using llama.cpp a while back, would be curious to see the speedup with this model.

https://github.com/TugdualKerjan/bunny/tree/main

The closest thing I believe we have as a neutral social media platform is Telegram. You can follow people, talk to friends with a performant interface without having a company who's main goal financially is making you see ads.

There is very little parasocial relationship on the platform although this could definetly grow as channels become a medium for influencers.

Do you think that this could be linked to a decentralized system for paying people to do this ? Similar to bitcoin, when a node goes down technicians arrive and are payed after the node has been fixed. (I see a lot of potential problems of "measuring" how much the node is repaired, who pays for it etc though)