anonymousDan

9 ms·

91.
by anonymousDan 1y ago
Why is that huge out of interest?
92.
by anonymousDan 1y ago
Not sure I agree with either/or. In person assessments are still pretty robust. I think an ideal university will teach both with a clear division between them (e.g. whether a particular assessment or module allows AI). What I'm cu
93.
by anonymousDan 1y ago
Is there any well known tool that can be used to scan an email account for PII (like the tools we have for scanning for confidential info in GitHub repos)?
94.
by anonymousDan 1y ago
No, the opposite!
95.
by anonymousDan 1y ago
As if the powers that be at YCombinator wouldn't roll over for Trump like all the other tech quislings.
96.
by anonymousDan 1y ago
How does this differ from Keystone?
97.
by anonymousDan 1y ago
A great example of the value of academia and scientific research.
98.
by anonymousDan 1y ago
Memory encryption typically doesn't keep anything encrypted within the CPU (e.g. in caches). Haven't looked at the details of this bug but I expect it wouldn't help for that reason.
99.
by anonymousDan 1y ago
What would be a typical/recommended server setup for using this for RAG? Would you typically have a separate server for the GPUs and the DB itself?
100.
by anonymousDan 1y ago
I don't agree with this. ATC is a good systems conference.
101.
by anonymousDan 1y ago
I think ATC is viewed as being a prestigious conference at roughly the level of some of the above.
102.
by anonymousDan 1y ago
> Today we have a huge oversupply of scientists, however there's too many of them to allow judging for potential, and many are not actually capable of dramatic impact. This is a fairly sweeping anti-science statement without any sup
103.
by anonymousDan 1y ago
So people with a Latin background are not part of the public?
104.
by anonymousDan 1y ago
Thanks - many of those didn't show up on your original link.
105.
by anonymousDan 1y ago
About 2 of those are actual exploits?
106.
by anonymousDan 1y ago
Pretty incoherent article. Not sure what point they are trying to make about the threat model of SGX. SGX was/is a groundbreaking attempt to solve a very difficult problem IMO. TEEs are still an active area of research that has benefit
107.
by anonymousDan 1y ago
The hype around agent protocols reminds me of the emperor's new clothes. There's just nothing to it from a technical perspective.
108.
by anonymousDan 1y ago
Sorry but this is such a shallow comment. In what way is the US government directing public funding to academic institutions not state control? It's just a different organisational framework that appears to have been more successful.
109.
by anonymousDan 2y ago
Half the global population lives on less than $7 a day. I admire your optimism, but it's perfectly possible to keep large swathes of people living in abject poverty.
110.
by anonymousDan 2y ago
Who actually knows? Far beyond what a UK university can afford.
111.
by anonymousDan 2y ago
Is there an equivalent of LORA using RL instead of supervised fine tuning? In other words, if RL is so important, is there some way for me as an end user to improve a SOTA model with RL using my own data (i.e. without access to the resource
112.
by anonymousDan 2y ago
It's well written but completely unjustified in its criticism of UK universities or their role given the resources required to train SOTA models. Are any US universities training SOTA models? No. Your point about the need for private v
113.
by anonymousDan 2y ago
What an utter bullshit take. As if the ATI could ever be competitive at LLMs given the resource requirements needed to train foundational models. If we want UK universities to continue to make contributions in this sphere we either need to
114.
by anonymousDan 2y ago
This sounds like a security nightmare.
115.
by anonymousDan 2y ago
Spoiler alert - they won't.
116.
by anonymousDan 2y ago
It's interesting that none of these papers seem to be coming out of academic labs....
117.
by anonymousDan 2y ago
Just to be clear, you're perfectly fine with this?
118.
by anonymousDan 2y ago
Yeah these are the kind of subtle issues I expected the article might be either glossing over or unaware of.
119.
by anonymousDan 2y ago
Presumably if you disable this on a per-transaction basis you are still guaranteed that transactions are replicated in FIFO order? Otherwise you could end up in weird situations where a synced txn reads from a lost txn and your state is eff
120.
by anonymousDan 2y ago
Is that ack sent once the request is received or once it is stored on the remote disk?

More ›