dpark

12 ms·

271.
by dpark 4mo ago
But only in the same sense that triangles are bad at representing curves, right? It seems that’s a wash.
272.
by dpark 4mo ago
> LLMs routinely hallucinate, or get "AI strokes" or get obsessed about not talking about goblins, etc. This isn't typical language behavior for humans unless they have severe neurological or psychological impairment. Pe
273.
by dpark 4mo ago
That doesn’t make any sense. A alphabet is a list of valid characters. A dictionary is not just a list. Even in a language like Chinese where individual characters carry meaning, a dictionary tells you what that meaning is . It’s not just
274.
by dpark 4mo ago
A tokenizer is not a dictionary any more than an alphabet is a dictionary.
275.
by dpark 4mo ago
Eh, I’ve seen it. I’m not entirely sure it’s entirely wrong either. Humans are certainly more than just next token predictors but it’s not clear that our typical language behavior is significantly different. We call it “stream of consciou
276.
by dpark 4mo ago
This is kind of a like assuming someone with bad spelling is stupid. Counting letters in a word seems to have little to do with understanding the word. Young kids can’t spell or count well at all but no one says that means they can’t unders
277.
by dpark 4mo ago
You’re still misinterpreting what I proposed. I didn’t propose a +0.5 debias at all. I proposed that the bias is removed by scaling. In this case you divide by 255.0 but multiply by 256. But again, the likely reason no one does this is beca
278.
by dpark 5mo ago
Kind of. According to the article, the mid-riser vs mid-tread distinction correlates to where the 0.5 is applied in the transform. I’m proposing that there is no 0.5 applied at all. Instead of counteracting the compression of the truncation
279.
by dpark 5mo ago
The entire issue arises from the use of truncation, right? It guarantees that only an exact 1.0 could land in the 255 bin so the net effect is a reduction of 256 bins to 255 bins. (Using random numbers as shown also guarantees no 1.0.) Why
280.
by dpark 5mo ago
We don’t know “what we are seeing” because we are looking from the outside. That’s my point. We can see a chat bot and we can see bad behavior and there are clearly a lot of assumptions that the problem is that someone gave the bot a set of
281.
by dpark 5mo ago
Evidence that it was actually AI based logic and not just a chatbot interface sitting on top of a shitty design.
282.
by dpark 5mo ago
Maybe? I don’t know what logic was actually in the LLM vs it just using a bad tool. Unless I missed it, the article had no actual context on that either. This looks like a terrible design rather than an AI problem to me, though.
283.
by dpark 5mo ago
> bad compatibility I’m curious what this means. Bad compatibility with Windows software? Or bad compatibility with Linux?
284.
by dpark 5mo ago
> an ARM device that had the battery life of Apple Silicon, yet was a real computer that wasn't locked down, ensured/promoted ARM compatibility with their ecosystem Isn’t that what this is? (Or is supposed to be?)
285.
by dpark 5mo ago
I had to go through the account recovery on my Facebook account once and the proof they demanded was that I match a bunch of pictures of friends to their names. I think it took 3 tries over multiple days to actually get it unlocked because
286.
by dpark 5mo ago
This exploit has essentially nothing to do with AI and everything to do with a terribly designed account recovery flow. This exact same flow could have been (and may have been; I don’t know how much the chatbot here actually does) staticall
287.
by dpark 5mo ago
> the managers who can't hire a ton angle for more senior people instead This is a huge factor. I’ve seen teams admit that they are too “senior heavy” and then still hire senior engineers over junior when the rare position opens u
288.
by dpark 5mo ago
You are assuming the assignment is reasonable and the candidate is lacking if they cannot complete in the expected time. And for all I know that’s entirely true for you. What I know is that I’ve seen assignments from others where the assign
289.
by dpark 5mo ago
So brave to create a new account to claim that Google employees should be executed. So brave.
290.
by dpark 5mo ago
This is theoretically true but it’s also rife with misaligned priorities. The people putting together these take home assignments have little incentive to ensure that they can be completed by a competent engineer in the allotted time. The e
291.
by dpark 5mo ago
I think a very real problem is that these take home problems are often way more than 4 hours. And to that they often add the traditional 4-6 hour interview loop. Provisional employment doesn’t work for most cases, though. It might attract p
292.
by dpark 5mo ago
I like Paint.NET. It’s a valuable resource. But the author is complicit in the spammy garbage. Turn off Adblock and try to download it. The official PDN website has multiple giant fake “download” ads. Then if you manage to click the real do
293.
by dpark 5mo ago
You are literally saying it’s a good idea to attempt to entrap potential employees by intentionally behaving the way you want them to not . It’s like pouring margaritas for you and the interviewee and then after they take a sip you say yo
294.
by dpark 5mo ago
Citation? They were contemporaries who knew of each other and apparently respected each other, including having students cross train across the test disciplines. I didn’t find anything saying they were military buddies.
295.
by dpark 5mo ago
> Aikido demonstration is cooperative. Uke and nage. Somebody throws and somebody receives the throw. I'm not one of those "Aikido is fake" people, but Aikido has no tournaments and isn't represented in MMA for a rea
296.
by dpark 5mo ago
I am glad you stopped asking that question. We all make mistakes and learn from them. An I correct in assuming the “4 letter word” was intended to make the candidate think of curse words? I’m curious what the intent of this question was if
297.
by dpark 5mo ago
Literally he’s saying to behave inappropriately in a professional interview and see if the candidate plays along. Might as well see if you can get the candidate to offer a bribe or sexual favors for the job since we’re going all in on entra
298.
by dpark 5mo ago
It’s absolutely not. Putting people in a vulnerable position and then pressing them for information you don’t need and should not ask for is a good way to demonstrate that you are an unethical or at best incompetent interviewer. It might be
299.
by dpark 5mo ago
> tell me your favorite 4 letter word Stop
300.
by dpark 5mo ago
Rejecting the company that attempts to put you through this kind of interview is how you pass. Remember that interviews are 2-way. You don’t have to engage in someone’s bad faith or incompetent interviewing.

More ›