Hacker Newsnew | past | comments | ask | show | jobs | submit | dragoncrab's commentslogin

How do LLMs calculate statistic metrics like average or standard deviation accurately in such experiments?


The background of the multiple hacks that led to the demise of Mt. Gox got a nice coverage on Darknet Diaries in 2017: https://darknetdiaries.com/transcript/9/


Since Monday numerous customers in Hungary are getting charged by Apple Pay without any reason. It looks that settled past trnsactions are booked again by error. Even blocking your debit card does not prevent the transactions. More coverage in Hungarian: https://telex.hu/techtud/2024/06/28/apple-bank-mnb-bankszove...


Totally agree.

I always cry on the inside when I see those beautifully designed, symmetrical, detailed sets taken apart and tossed together to form the next ninja castle or whatever, but after all Lego is to be taken apart and my 6-8-9 year olds have no remorse in doing so.


And this means a 10x increase in the thermistor price compared to the toaster. It cold be even as high as $2 for a piece!


I don't want to be rude, but given the average quality of material produced by big consulting companies I have encountered over my career, I'm not that surprised that GPT4 beats them.


All the big corps I ever worked for had their devices configured to go through the central VPN and proxy even for browsing.

But I guess blocking $$$ corporate visitors should be fine by you as well.


The problem with VPNs is the exit node and not the act of using a VPN in itself. You will also be fine if you VPN to a clean IP, which is why corporate VPNs aren't generally affected. The problem is that for better or worse, public VPN exit nodes have a history of abusive activity and are blocked as a result.


It is extremely easy to differentiate between the two with a commercial VPN detection product rather than just someone's hobby proof of concept.


Have visited the site through my self hosted OpenVPN tunnel and got 0 for both the VPN and proxy score.


Same here, although it caught the browser clock differential which I forgot about.


The title is misleading. Only the fairly new and surprisingly expensive SMRs were taken into account, the much cheaper traditional nuclear power plants weren't. Which could be fine, but then don't generalise your findings across the entire nuclear landscape.


The economics of traditional plants are even worse. They're complicated to build, they take forever to throttle, they need a ridiculous amount of auxillary cooling infrastructure due to crap throttling, and their thermal efficiency is below even fossil fuels since they can't run superheated without making the reactor even more of a maintenance nightmare


Given how easy it would be to turn in an AI generated patent claiming that you were the inventor I think this is some legal activism and the guy doesn't actually care about these patents.

I can't tell the goal, but is likely about clarifying the legal framework or even influence it by creating precedents in these clear scenarios.


The person filing the patent has a lot of prior work in trying to get AI stuff legally recognised: https://www.wired.co.uk/article/stephen-thaler-ai-copyright-...


Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: