porridgeraisin

7 ms·

1.
by porridgeraisin 5d ago
Well yeah. If for some task you find it easier to just do it yourself then you should. But you can improve the situation even if you can't entirely automate it by automating parts of the verification thereby making it easier to human-d
2.
An Age of Experimentation [pdf](thomasdullien.github.io)
1 points by porridgeraisin 5d ago|0 comments
3.
by porridgeraisin 5d ago
Well, running an LLM X amount of times does give you better results provided you are willing to select the best one out of the X yourself. But I agree with your general point. One of the reasons subscription plans are cheaper because they m
4.
by porridgeraisin 5d ago
It gets very hard after childhood. This kid I know has learnt russian, chinese, arabic, spanish, hindi, apart from english and his mother tongue tamil. All with dedicated native teachers and zoom classes, where he actually speaks to a bunch
5.
by porridgeraisin 6d ago
https://archive.is/cE2XZ
6.
India Outsourcing Shifts Upmarket as AI Reshapes Jobs, ING Says(bloomberg.com)
2 points by porridgeraisin 6d ago|1 comments
7.
by porridgeraisin 6d ago
I think their problem is more not being cited by the team at typesafe, as in general academic politeness. On the one hand you have the charitable assumption that they developed it independently. On the other hand, my opinion is that it is n
8.
Guest to host: escaping Docker's hypervisor(accomplish.ai)
4 points by porridgeraisin 7d ago|0 comments
9.
Much Ado About Noising: Dispelling the Myths of Generative Robotic Control(arxiv.org)
1 points by porridgeraisin 8d ago|0 comments
10.
Probably – a programming language for LLM workflows, powered by Jev(probably-lang.southpolesteve.workers.dev)
3 points by porridgeraisin 8d ago|1 comments
11.
by porridgeraisin 9d ago
In RLCD (which is now an RL acronym that has 3 different unrelated expansions!), you basically massively negatively reward a distribution that is {yes: 0.9, no: 0.1} if the answer was no, and less negatively reward a {yes: 0.6, no: 0.4}. Ma
12.
A DuckDB extension where you can use Jev to do quick classification of rows(twitter.com)
1 points by porridgeraisin 9d ago|0 comments
13.
by porridgeraisin 10d ago
Yea. I have asked it to verify my ideas with experiments sometimes. And it cheats and warps the results so that the results are reached
14.
by porridgeraisin 10d ago
Yes, when post training models for long tasks this happens gradually. It is not easy to prevent it as such.
15.
by porridgeraisin 11d ago
Yep. That's exactly the case. External graph state/rudimentary planner + LLM proposer + cheap verifier gets so much done.
16.
by porridgeraisin 11d ago
My bad
17.
by porridgeraisin 11d ago
No, you did not read what I wrote properly. What I wrote clearly is that we _want_ the LLM to do tasks on the full global open internet. That is the whole point of the technology. So airgapping or any form of total sandboxing is not on the
18.
by porridgeraisin 11d ago
Are you serious? Just because you have nuclear weapons doesn't mean all other forms of military are useless. I don't even know how to parse that argument to respond in more detail.
19.
Randomized query complexity can beat certificate complexity(arxiv.org)
2 points by porridgeraisin 11d ago|0 comments
20.
by porridgeraisin 12d ago
That's a bit pedantic no. Memorization in ML refers to the model having the wrong level of capacity such that it's too hard to optimise it such that it doesn't memorize the _training examples_ themselves.
21.
by porridgeraisin 12d ago
I literally address that in the last sentence
22.
Rapidly scaling online storage to serve over 1B ChatGPT users(openai.com)
2 points by porridgeraisin 13d ago|0 comments
23.
by porridgeraisin 13d ago
26ghz mm wave was a bungle as well. See https://news.ycombinator.com/item?id=49675724
24.
How to Build a Custom Agent Harness(langchain.com)
2 points by porridgeraisin 13d ago|0 comments
25.
by porridgeraisin 13d ago
Come on, yoshua bengio of all people knows how post training works. While I too don't like anthropomorphisation, I would give it a more nuanced reading. His point is that today we are giving it reward to complete the task, and it may t
26.
by porridgeraisin 13d ago
It's not chromium. But what's happening here is that even if you don't click paste, zoom is actively listening to clipboard events and consuming pastes. Tbh, if they aren't harvesting clipbakrds data which is a weird thi
27.
by porridgeraisin 13d ago
https://x.com/badlogicgames/status/2098868942685012405
28.
An Advanced System Architecture Breakdown of OpenAI's Jalapeno Accelerator(siliconcodesign.com)
8 points by porridgeraisin 13d ago|0 comments
29.
by porridgeraisin 14d ago
There was full team that found out that did the test that 26ghz doesn't work IRL. It didn't even work across a street corner. But they were ignored in favour of some well, career pushers. In india, jio is using 26ghz for wifi now
30.
by porridgeraisin 14d ago
Yeah. In india jio was SA from the start. Airtel began NSA then I think now they have mostly moved to SA.

More ›