inkysigma

9 ms·

61.
by inkysigma 11mo ago
It's unlikely the specific codec that is the issue but the bug report suggests that the code path could be hit by a maliciously crafted payload since ffmpeg does file fuzzing. They almost certainly have ffmpeg stuff that touches user s
62.
by inkysigma 11mo ago
The article is a bit sparse of technical details but am I misunderstanding what they're doing or are they describing a field that's isomorphic to C but described as a pair of real numbers? If so, I don't see how that meaningf
63.
by inkysigma 11mo ago
I don't think that's an accurate description of the full scope of the problem. The codec itself is mostly unused but the code path can possibly be triggered from file fuzzing that ffmpeg uses so a maliciously crafted payload (e.g.
64.
by inkysigma 11mo ago
Not being told the existence of bugs is different from having a warranty on software. How would you submit a patch on a bug you were not aware of? Google should provide a fix but it's been standard to disclose a bug after a fixed time
65.
by inkysigma 11mo ago
Maybe for a small project? I think the difference here is rather minimal. Everybody "knows" code often has security bugs so this announcement wouldn't technically be new information. For a large project such as ffmpeg, I doub
66.
by inkysigma 1y ago
Sorry, is this person comparing the rights and immunities of a head of a sovereign nation to those of a CEO of a company? I don't think France, as a sovereign country, is completely bound by US law whereas binance, when it is operating
67.
by inkysigma 1y ago
Maybe, but then this seems more like an exercise in distillation rather than solving the original problem which is what the title "Folding proteins is simpler..." suggested to me at least. Part of the problem with any ML task is t
68.
by inkysigma 1y ago
I think you're selling this too far with "one of the more ethical tech companies around" and "a force for good". You'll have to clarify what exactly that comparison is based on. I'm not a total fan of Appl
69.
by inkysigma 3y ago
> You really think you are going to ask a child what are they doing every second and what they talked about and they are going to tell the truth? Of course not, but usually adults can force a passcode out (or take the device altogether)
70.
by inkysigma 3y ago
In the US, where the case was decided, it's about 50%.
71.
by inkysigma 3y ago
It's a difficult security model when the threat actor is a parent who presumably has access to the device and in an unlocked state along with permissions to install anything from anywhere. It's not like the threat actor realistica
72.
by inkysigma 3y ago
As a quick point of clarification, I don't think MAMBA has a convolutional view since it drops the time invariance and is strictly linear. The authors use parallelized prefix sum to achieve some good speed up.
73.
by inkysigma 3y ago
Do you also exclude contractors that don't have a fully decked out Ford F-350? What if they carry a cheap $1 pen instead of a $50 pen to write some quotes for you? Also, you realize there are android phones that are plenty expensive ri
74.
by inkysigma 3y ago
Without a search engine, how am I supposed to weigh the accuracy of an LLM? How am I supposed to take responsibility for ensuring accuracy? I also think people who say that search engines lie are seriously overestimating the amount of lies
75.
by inkysigma 4y ago
> This is not surprising. A human would suffer from similar errors at a similar rate if it were exclusively fed an interpretation of reality that only consisted of text from the internet. I think this is surprising at least if the bot ac
76.
by inkysigma 4y ago
Like yes to some extent there is a mild amount of generalization in that it is not literally regurgitating the internet and it to some extent mixes text really well but I don't think that's obviously the full on generalization of
77.
by inkysigma 4y ago
I have to disagree with that. Like maybe for a toy example to demonstrate what I'm talking about, imagine I was teaching you the addition operation mod 100 and I gave a description of the operation f(x, y) = x + y % 100 for x, y in Z_1
78.
by inkysigma 4y ago
It's almost nearly always wrong in my experience. Like if I ask it a basic technical question, it gets stuff mostly correct but if you look too much at the details, then errors start popping up. If you ask it something relatively compl
79.
by inkysigma 4y ago
How exactly can I tell if the chat bot is hallucinating or not without actually going into the search result source (at which point the bot becomes less useful)? It's hallucinating what the sources are saying and making up what the sou