lucidrains

trying to bring verification to bio? it is going to be tough, there's just an unbelievable amount of noise in wet lab work

there is a 2018 documentary on her life titled Kusama: Infinity. it appears you can watch it online for free now.

Reminiscent of his other short stories from the Martian Chronicles, also rife with apocalyptic themes

quite surprised to see SAC, considering the deepmind ping pong paper resorted to evolutionary strategies, iirc

They need to have a service that allows for pets. It is the only reason I still use Ubers!

there has been some movement on that front, in the form of adding expand / delete tokens! https://hkunlp.github.io/blog/2025/dreamon/

It is a reference to the beatles song, mainly because Noam Shazeer is a music lover

very nice, will have to try it out! this is the same research group from which Robin Rombach (of stable diffusion fame) originated from

I am sorry to hear that.

indeed, it just becomes less likely

if you are a man and make it past age of 29 without starting to hear voices, you can breathe a sigh of relief (I did)

Love to know as well!

oh wow, I was looking for use of islands or map-elites that I missed this.. thought it was the blandest mimetic evolution possible

does this not mean we should explore usage of talking heads (Shazeer et al) a bit more? https://arxiv.org/abs/2003.02436

Indeed, the best setting is in nature on a beautiful day

amazing work! thank you!

would be interested to see thunderkittens (great name!) tackle the flash attention backwards pass, which is an order of magnitude harder than the forward

wow, so prophet net does work! i spent so much time experimenting with it back in the day, but just lacked the scale to see a positive result.

indeed, simsiam is a great example of the effectiveness of using stop gradient

could even try it with a fraction of the attention heads, instead of introducing new tokens

yes, it is a stop gradient mask on the attention matrix, iiuc. worth trying

I can't think of anyone better to teach attention mechanism to the masses. This is a dream come true

I've never worked with a project with more than 7 dimensions, yet

oh interesting, Jeff Hawkin's HTM?

"There are only two tragedies in life: one is not getting what one wants, and the other is getting it." - Oscar Wilde

actually got some government shill who can't even spell to respond. touch a nerve?

exactly, this guy understands how business in China works.

would be nice if it were just more attention. there could be something else though