krupan

20 ms·

541.
by krupan 11mo ago
"Would you trust a medical system measured by: which doctor would the average Internet user vote for?" Isn't this exactly what we do when we consult online reviews for everything?
542.
by krupan 11mo ago
"Now, with LLM-generated content, it’s hard to even build mental models for what might go wrong, because there’s such a long tail of possible errors. An LLM-generated literature review might cite the right people but hallucinate the pa
543.
by krupan 11mo ago
I'm trying to understand where this kind of thinking comes from. I'm not trying to belittle you, I sincerely want to know: Are you aware that everyone writing software has the goal of releasing software so perfect it never needs
544.
by krupan 11mo ago
I love how LLMs have made everyone forget how hard it is to verify software correctness and how hard it is to maintain existing software. There is endless gushing about how quickly LLMs can write code. Whenever I point out the LLMs make a
545.
Unix v4 tape found in closet at UofU(ksltv.com)
9 points by krupan 11mo ago|1 comments
546.
by krupan 11mo ago
Does that cout the dams that flood valleys and displace thousands of people, plants, and animals from their homes?
547.
by krupan 11mo ago
People outside the 4 U.S. Timezones exist?
548.
by krupan 11mo ago
I love how everyone thinks software is easy to validate now. Like seriously, do you have any awareness at all about how much is invested in testing software by the likes of Microsoft, the game studios, and any other serious producers of so
549.
by krupan 11mo ago
No mention of Bitcoin. Exploiting ethereum smart contracts is nothing that new or exciting.
550.
by krupan 11mo ago
TFA's while point is that there is no easy way to tell if LLM output is correct or not. Driving mistakes provide instant feedback if the output of whatever AI is driving is correct or not. Bad comparison.
551.
by krupan 11mo ago
"I feel like these distinctions are equivalent to LLMs can make mistakes. Humans can't." No, that's not it. The difference between humans and AI is that AI suffers no embarrassment or shame when it makes mistake
552.
by krupan 11mo ago
I've been using emacs in terminal mode inside termux for a few years and it's not bad. Full GUI emacs would be nice, I'll have to give this a try
553.
by krupan 11mo ago
There is no life without pain and suffering. It's as much a law of nature as gravity. Life needs something to push against, to struggle to overcome. Without that there simply is no life. If life doesn't struggle and suffer aga
554.
by krupan 11mo ago
Writing, journaling, is such a good way to process thoughts and emotions and this is a great example. I'm grateful for this person sharing her very personal journal entries with us even though it is not required at all. It sets a gre
555.
by krupan 1y ago
"ask the developers to practice discipline" is a baseline requirement for coding in C
556.
by krupan 1y ago
Not sure why this is down voted when the whole point of TFA is to torture the C language into doing something it can't really do. I guess there's an unspoken assumption in TFA that you are stuck using C and absolutely cannot use
557.
by krupan 1y ago
You should probably ask an AI to read it and summarize it for you.
558.
by krupan 1y ago
Someone should have kept investing in Cruise for that. They were basically on par with Waymo, but then GM stopped the funding and shut Cruise down
559.
by krupan 1y ago
Man I hope you're right
560.
by krupan 1y ago
This is by side.hustlereview on Instagram. He tries the side hustles that others on Instagram are selling and makes reels reviewing them. Unsurprisingly, most do not live up to the hype.
561.
Reviews for Online Courses(review.courses)
1 points by krupan 1y ago|1 comments
562.
by krupan 1y ago
This type of announcement is definitely pumping air into it
563.
by krupan 1y ago
The author appears to be Chinese and the articles they have written on the site have something to do with China
564.
by krupan 1y ago
Your "in theory" is doing a lot of work there. So much work. Have you heard about no-fly lists? The latest ICE actions? The Red Scare? Giving the government MORE power is almost never the answer.
565.
by krupan 1y ago
Why didn't the people selling shovels to gold miners dig for gold themselves? Because Nvidia is making actual profit selling hardware to those who do, not hoping for a big payout sometime in the future. Different risk/reward mode
566.
by krupan 1y ago
Before your edit I was going to answer, sadly no, they can't even be repurposed for Bitcoin mining.
567.
by krupan 1y ago
Do they hallucinate more? How many words per day do humans spit out vs the LLMs, and what is the hallucination rate? Also, don't count humans saying, "I'm not sure but I think..." as a hallucination, because it's
568.
by krupan 1y ago
Um, no. They are unreliable at a much faster pace and larger scale than any human. They are more confident while being unreliable than most humans (politicians and other bullshitters aside, most humans admit when they aren't sure abo
569.
by krupan 1y ago
Sounds like these two engineers really pissed these black hat hackers off! If that doesn't boost their resumes then I don't know what will!
570.
by krupan 1y ago
Magazines and such had a lot of ads too

More ›