Y
HN Search
Hacker News Search
new
|
comments
|
top
|
jobs
wseqyrku
searching PlanetScale…
1.
▲
2.
▲
3.
▲
4.
▲
5.
▲
6.
▲
20 ms
·
391.
▲
by
wseqyrku
3y ago
They both are all about publicity and that's the name of the game. Doesn't matter who wins at the end, it's going to be absolutely all out in the Open. Hey.
392.
▲
Alibaba Presents EMO: Emote Portrait Alive
(twitter.com)
1 points
by
wseqyrku
3y ago
|
0 comments
393.
▲
Pluralistic Ignorance
(en.wikipedia.org)
3 points
by
wseqyrku
3y ago
|
1 comments
394.
▲
by
wseqyrku
3y ago
Is there anyone on HN to comment on this?
395.
▲
What would you attempt if you knew P equals NP?
3 points
by
wseqyrku
3y ago
|
6 comments
396.
▲
by
wseqyrku
3y ago
That doesn't even compile in English.
397.
▲
by
wseqyrku
3y ago
Feeding the output to another inference would probably reduce hallucination? but I have this impression that the models talking to each other don't need to know English, a byte stream would do. Just like the hidden layers of an ANN.
398.
▲
by
wseqyrku
3y ago
So they always hallucinate, it's just sometimes good enough?
399.
▲
by
wseqyrku
3y ago
I've seen this one from Google which is kind of peculiar. Your client does not have permission to get URL / from this server. _That’s all we know._
400.
▲
by
wseqyrku
3y ago
> How to hire low experience, high potential people So basically, how to hire best for cheap.
401.
▲
by
wseqyrku
3y ago
Oh that's why no one's looking when they're making a turn. All the incredibly stupid car accident clips I've ever seen make sense now.
402.
▲
by
wseqyrku
3y ago
You're looking at it just like how Microsoft did, either way 2000 people lost their jobs. And I'm down voted for some reason.
403.
▲
by
wseqyrku
3y ago
> 2k employees look like a lot, 8% looks kinda better. Sorry what?
404.
▲
by
wseqyrku
3y ago
This seems to work fine for Temporalio.
405.
▲
by
wseqyrku
3y ago
Has the US downgraded to a third world shit hole that people/companies tend to be concerned with this kind of stuff? Disclaimer: I live in such a place and this is normal.
406.
▲
by
wseqyrku
3y ago
You guys really expected that military would sit on the corner and keep saying killer robots bad?
407.
▲
by
wseqyrku
3y ago
If MS could acquire Discord back in 2021 people would assert that this is the direct result of the acquisition.
408.
▲
Improving Reasoning in Language Models with Layer-Selective Rank Reduction
(arxiv.org)
2 points
by
wseqyrku
3y ago
|
0 comments
409.
▲
by
wseqyrku
3y ago
Sounds like 2024 is gonna be the year of lawsuits like this.
410.
▲
New York Times Sues Microsoft and OpenAI for 'Billions'
(bbc.com)
3 points
by
wseqyrku
3y ago
|
1 comments
411.
▲
by
wseqyrku
3y ago
> The more rational voices in my mind, though, become more and more afraid of a world where the only thing you can trust is people sitting right in front of you. You could care less if the content you already consume on forums like HN we
412.
▲
SparQ Attention: Bandwidth-Efficient LLM Inference
(arxiv.org)
1 points
by
wseqyrku
3y ago
|
0 comments
413.
▲
by
wseqyrku
3y ago
Please, look it up https://pika.art
414.
▲
by
wseqyrku
3y ago
you haven't seen pika then.
415.
▲
by
wseqyrku
3y ago
So far you were a line in the log. Now someone is actually looking at you with three eyes.
416.
▲
Exploring an LM to Generate Prolog Predicates from Mathematics Questions
(paperswithcode.com)
5 points
by
wseqyrku
3y ago
|
0 comments
417.
▲
No Feature
(ia.net)
2 points
by
wseqyrku
3y ago
|
0 comments
418.
▲
Wikipedia: Removal of Sam Altman from OpenAI
(en.wikipedia.org)
1 points
by
wseqyrku
3y ago
|
0 comments
419.
▲
Do Large Language Models Know What They Don't Know?
(arxiv.org)
2 points
by
wseqyrku
3y ago
|
0 comments
420.
▲
Kubernetes Case Study: OpenAI
(kubernetes.io)
1 points
by
wseqyrku
3y ago
|
0 comments
More ›