Y
HN Search
Hacker News Search
new
|
comments
|
top
|
jobs
brantmv
searching Neon…
1.
▲
2.
▲
3.
▲
4.
▲
5.
▲
6.
▲
5 ms
·
1.
▲
by
brantmv
6mo ago
Very nice. But surely the most chopped up amen break is Virtual Riot's Death by Amen . https://youtu.be/XpnNVWOC98A
2.
▲
We Made the Isospectral Drums
(prismika.github.io)
2 points
by
brantmv
7mo ago
|
1 comments
3.
▲
by
brantmv
7mo ago
Maybe you've heard of the "can one hear the shape of a drum" question. In the 90's, mathematicians successfully produced two drum shapes that would theoretically produce indistinguishable sets of frequencies when struck.
4.
▲
by
brantmv
9mo ago
For anyone wondering, "mse" probably means "math Stack Exchange".
5.
▲
by
brantmv
9mo ago
Also, the mathematical content here is pretty thin. Their main theorem has nothing to do with LLMs directly. It's a theorem about a five-state Markov chain, and the proof follows from standard Markov chain theory. For those reasons, th
6.
▲
by
brantmv
9mo ago
Maybe I'm wrong, but it looks like the authors did not actually have any LLMs write or verify any code for their experiments. Instead, their experiments consist of simulating the simplified Markov chain model itself . They simulated t
7.
▲
by
brantmv
9mo ago
Why say this in such a rude way?
8.
▲
by
brantmv
11mo ago
I think the surprising part is not that the necessary number of poisoned documents is small , but that it is small and constant . The typical heuristic is that a little bad data is not so bad; if you have enough good data, it'll all
9.
▲
by
brantmv
1y ago
Mathematicians are afraid of higher order tensors because they are unruly monsters. There's a whole workshop of useful matrix tools. Decompositions, spectral theory, etc. These tools really break down when you generalize them to k-tens
10.
▲
by
brantmv
1y ago
I'm always surprised how many other mathematicians don't know what I'm talking about when I reference this paper. It should be in the canon of math essays.