pizza

22 ms·

1.
by pizza 20d ago
They’ve provided insane value to the ai community. Imo is there really any obvious way to standardize model/preprocessing/inference/training/rl recipes across the space of frankenstein model family/architecture&#x2F
2.
by pizza 1mo ago
Compression is just counting. Probability, also, pretty much, just counting. For these reasons I think the role of information theory in describing the process of the development of reasoning and the gain of understanding has been overstate
3.
by pizza 4mo ago
you can say the same thing of the watts in a person too
4.
by pizza 4mo ago
Had to butcher the title slightly to get it under the limit- original: Compression And Decompression Under FHE Using Error-Correcting Codes and Copy-And-Recurse
5.
Compression & Decompression w/ FHE via Err Correcting Codes and Copy-and-Recurse(eprint.iacr.org)
1 points by pizza 4mo ago|1 comments
6.
Learning Pseudorandom Numbers with Transformers(arxiv.org)
11 points by pizza 5mo ago|3 comments
7.
Objects of Categories as Complex Numbers (2002)(arxiv.org)
2 points by pizza 5mo ago|0 comments
8.
To Dissect a Mockingbird: A Graphical Notation for the Lambda Calculus (1996)(dkeenan.com)
11 points by pizza 5mo ago|1 comments
9.
by pizza 5mo ago
For most tasks, at some future date, isn't there going to be some ambient baseline of capabilities you can get per $/tok, starting at ~0 for OSS models, such that eventually all tooling gets trivially transferable?
10.
All elementary functions from a single binary operator(arxiv.org)
858 points by pizza 5mo ago|298 comments
11.
by pizza 5mo ago
OP is correct; surprisal is outcome-dependent and entropy is distribution-dependent - entropy is E_p[informativeness of measuring outcome x] - take n outcomes, then a distribution over them lives on the simplex \delta ^ (n - 1). you can lif
12.
by pizza 6mo ago
I meant in the sense of - you have benchmarkers and trainers. If you publicize your evaluation, trainers may likely have their models 'consume' it, even if only indirectly: another person creating their own benchmark from scratch
13.
by pizza 6mo ago
There’s a Dark Forest problem for evals. As soon as they’re made public they start running out of time to be useful. It’s also not clear how to predict how the model will perform on a task based on an eval. Or even whether, given two skills
14.
by pizza 6mo ago
In Singapore it seems 80% of people live in public housing https://en.wikipedia.org/wiki/Public_housing_in_Singapore though I can't speak as to what the effect is on its housing market
15.
by pizza 6mo ago
I mean. Sounds like the guy had existing long term goals, needed to overcome an activation threshold, and used AI as a catalyst to just get started. Seems like, behaviorally, AI was pivotal for him to learn things, even if the things he lea
16.
by pizza 6mo ago
“never go to sea with two chronometers, take one or three”
17.
by pizza 7mo ago
here-doc usage has probably 100x-ed in the last year
18.
Levallois Technique(en.wikipedia.org)
4 points by pizza 7mo ago|0 comments
19.
Masakhane(masakhane.io)
2 points by pizza 7mo ago|0 comments
20.
by pizza 7mo ago
Why?
21.
CADRE: Cooperative Autonomous Distributed Robotic Exploration(jpl.nasa.gov)
1 points by pizza 7mo ago|0 comments
22.
by pizza 7mo ago
The more general question of how to evaluate the quality of a given skill file is quite interesting to me. A skill may prime a model's responses in a way that a prompt alone may not. But also models aren't good at judging what the
23.
by pizza 7mo ago
The possibility of intelligent machines undergoing transformative regeneration actually dates back to a party hosted by one Charles Babbage where, in attendance, was one Charles Darwin, who only thereafter published On the Origin of Specie
24.
DOE Genesis Mission List of National Science and Technology Challenges(energy.gov)
1 points by pizza 7mo ago|0 comments
25.
by pizza 7mo ago
I think you're mistaking the .wav as the final product, whereas instead it's really the .html blog post and this discussion.
26.
Surfaces with Klein bottle topology occur in fusion reactor fields(arxiv.org)
3 points by pizza 8mo ago|1 comments
27.
by pizza 8mo ago
This x thread may not be the best source of clarity on what is actually being default opted-into. Sorry. I looked into it and it seems that Starlink denies browsing history would be shared [0]. Seems I can't edit the title any more. &g
28.
Starlink users must opt out of all browsing data being used to train xAI models(twitter.com)
97 points by pizza 8mo ago|28 comments
29.
Restoring Locality:Heisenberg Picture as Separable Description of Quantum Theory(arxiv.org)
2 points by pizza 8mo ago|0 comments
30.
by pizza 8mo ago
That sounds really entitled.

More ›