So because of the way LLMs are trained they naturally prefer to use the language one has initiated the conversation. I would guess that for the LLM, or specifically the transformer in it, the difference between human languages is not that big. But it would be interesting what it would lead to if we give transformers a possibility to answer in much more denser spaces. Like what if we can extract concepts itself and allow the LLM to answer with them instead of using the human language intermediate. - think a compressed embedding
I cannot recommend FT.
I tried to get their physical paper, never got one and although I made 3 calls and wrote about 10 mails to them they only would escalate my ticket after I unsubscribed and demanded a refund, this was after 3 weeks and a lot of time on my side.
That is why I am also super interested in just printing news from the net for myself, so I do not need to keep watching on a screen.
Are you in the UK and within some short distance of civilisation? If so, you very likely have a newsagent near by you and there's a surprisingly good chance they still do a paper delivery round.
Obviously if you're in a hut up a mountain or live in Norfolk then this may be less useful advice for you.
Ah yes. Filling newspapers with petty leaflets of spam the night before and then having to get up at 6am in the cold rain to get lost cycling around an estate block smelling smells of god knows what only to be paid less than minimum wage. what fun times.
At least I could then afford to buy a packet of cigs, and then upsell them to the chavs for a higher price...
Ohhhh this is soo cool!
I always wondered if my color perception is normal because sometimes i have the feeling that i do not have that much of saturation.
Still failed to find such a test but this goes into this direction.
Maybe this comment can help me with this search.
This article is not about "allowing" kids to be bored.
It is to force kids to spend their time like this person that has a romanticized view of their childhood wants them to fill "the void".
I believe there is a point there, we live in a society where we must be entertained well! Not being entertained is shame. Allowing is fine enough word here I believe.
Rather then taking kinds their media away and in turn making this media even more valuable for them. I would suggest trying to establish attention as something valuable and that is important to direct yourself rather then having it directed by someone else.
This remembers me of NVIDIA RTX Voice [0].
Although not made to isolate single persons, this is quite impressive.
I hope that this single person isolation will find it's way to consumer noise-cancelling headphones
I think one could build quite the good system with 2 directional microphones and then do some beamforming or how it is called to isolate the depth one want to perceive.
But this is super expensive since you need calibrated mics etc.
The biggest advantage of neural nets in this field is that you can use a dirt cheap microphone and postprocess it so good that it is good enough or even very good for humans.
So because of the way LLMs are trained they naturally prefer to use the language one has initiated the conversation. I would guess that for the LLM, or specifically the transformer in it, the difference between human languages is not that big. But it would be interesting what it would lead to if we give transformers a possibility to answer in much more denser spaces. Like what if we can extract concepts itself and allow the LLM to answer with them instead of using the human language intermediate. - think a compressed embedding