- What bias?
It seems to me that in your view the sheer openness to evaluate LLM use, anecdotally or otherwise, is already a bias.
I don't see how that's sensible, given that to evaluate the utility of something, it's necessary to accept the possibility of that utility existing in the first place.
On the other hand, if this is not just me strawmanning you, your rejection of such a possibility is absolutely a bias, and it inhibits exploration.
To willfully conflate finding such an exploration illegitimate with the findings of someone who thinks otherwise as illegitimate, strikes me as extremely deceptive. I don't appreciate being forced to think with someone else's opinion covertly laundered in very much. And no, Tao's comments do not meet this same criteria, as his position is not covert, but explicit.
- > so are kings of England, Denmark, Sweden, Norway, Spain, Netherlands & Arabia
The first six of which to my understanding are token roles, sometimes de jure, but definitely de facto. This is unlike the role of the Iranian Supreme Leader, who according to my findings is heavily involved and has the license to do so.
I don't know what country Arabia is supposed to be.
> The Guardian Council is an elected body with the power to remove the so-called "Supreme Leader", so his power is limited too
You know that I considered and checked for all of this ahead of time, right? These loopholes are very on the nose.
The Supreme Leader directly elects 6 of the Guardian Council members, but also the judiciary chief, who elects the other 6.
The Supreme Leader is elected and kept in position by the people-elected Assembly of Experts, the applications for which are... filtered through by the Guardian Council. And considering they have never exercised their authority to as much as condemn but especially to remove a Supreme Leader according to my search, it's a reasonable conclusion that either everything is magically harmonious over there, or this authority in practice is a token one too.
I have also looked into whether these bodies are partial, and not only is this true covertly, it is true openly. Candidates that don't fit the bill on any of these levels are proudly filtered out in droves.
I'm really quite unconvinced you're here to inform with honesty in mind.
- > ChatGPT will happily do its own private "literature search" and then not tell you about it
Also known as model inference. This is not something "private" or secret [*]. AI models are lossily compressed data stores and will always will be. The model doesn't report on such "searches", because they are not actual searches driven by model output, but just the regular operation of the model driven by the inference engine used.
> even Terence Tao has freely admitted as much
Bit of a (willfully?) misleading way of saying they actively looked for it on a best effort basis, isn't it?
[*] A valid point of criticism would be that the training data is kept private for the proprietary models Tao and co. using, so source finding becomes a goose chase with no definitive end to it.
An I think valid counterpoint however is that if locating such literature content is so difficult for subject matter experts, then the model being able to "do so" in itself is a demonstration of value. Even if the model is not able to venture a backreference, by virtue of that not being an actual search.
This is reflected in many other walks of life too. One of my long held ideas regarding UX for example is that features users are not able to find "do not exist".
- Thank you, I'll try to keep it in mind. I'll admit that the curtness of my original question was not just you misreading it, but it did (also) come from a place of genuine confusion.
For what it's worth, it's not even that I don't see merit to their points. I'm just unable to trust that they're being genuine, not the least for how they conduct themselves (which I only fault them for so much). This also impacts my ability to reason about their points clearly.
Sadly, I'm not able to pitch any systematic solutions.
- What scientific field do you reckon the regular usage of LLMs falls under? Do you genuinely think Tao was making scientific claims or just provided evidence that may eventually feed into some? It reads to me like just a plain recollection of events, an anecdotal experience.
- It genuinely seemed to me that they were looking for empirical reproductions of a formal proof, which is a nonsensical demand and objection given what formal proofs are. My question was spurred on by this and genuine.
I now see in the other subthread what they mean.
- Do you know what a formal proof is?
- > The "regime" is a republic with regularly held presidential (8 presidents in 45 years) and parliamentary elections.
I'd think the regime thing refers to the Supreme Leader of 36 years and his Guardian Council, no?
- But anti-government protests you do acknowledge exist. What's preventing you from referencing those to demonstrate that they are small and few in comparison? What's preventing you from demonstrating that they're Israeli orchestrated like you're claiming?
> 100% of the people falsely claiming there are massive anti Iran protests are militant Zionists.
So your evidence is just you being 100% certain?
Edit:
In reply, you inquired if I myself was a Zionist, and if I thought Israel had a right to exist. Your post was flagged before I could formulate a response.
I'm not a Zionist. Not for any principled reason mind you, but because up until half an hour ago I did not even know what it was about exactly. Had to look it up.
Regarding your second question, I don't think countries come into existence by having some sort of (moral or legal) right to do so, so that's a bit of a loaded question I'm thus not able to answer.
I really don't know why I even bothered to answer these with honesty, you clearly did not ask in good faith. Should have just dodged them like you did with my questions.
- So could you please provide evidence for that? I specifically mean evidence that those other protests are:
- insignificant in size or are significantly less numerous;
- artificially orchestrated;
- Zionist in nature?
- It just references the same underlying event, which your own source reports to be a counter-protest, directly falsifying your claim.
- > The event was framed by authorities as a demonstration of national solidarity amid a backdrop of ongoing unrest and anti-government protests across multiple Iranian cities, which entered their second week
- People can focus on a lot of things and make any arbitrary narrative emerge. My problem is exactly that I do not find this angle compelling so far, especially in light of the to-me-obvious alternative.
You started off by listing a bunch of things that did not pass my smell test (and you have now walked back on), then followed it up by what's essentially a scattershot of vague gesturings. Why would I focus on what you tell me to? Not only is any of these not compelling, I do not find you a reliable narrator so far at all.
- I gave this a skim and a keyword search. Note that I'm not familiar with the matter.
The article claims that the Rohingya genocide in Myanmar that kicked off in 2017 has been substantively fueled by Facebook propaganda efforts, with strong links to Myanmar's own "security forces" (military).
> it's to prevent foreign interference (like CIA) from fueling civil unrest and spreading AI deepfakes, as seen in Myanmar and Brazil
In contrast then, you seem to allege that it was actually a foreign interference campaign by the CIA? Or am I misunderstanding what you're proposing?
Because if I'm not, I fail to see how what you linked supports that at all. Even your mention of deepfakes seems very questionable, as those haven't been a thing until late 2017, by which point this cleansing effort was already long underway. I further see that the US has formally condemned these events, although of course that does not rule out involvement.
- Anything that is both streamable and seekable?
- They're actually two separate claims, one of which the blogpost does support. The other one is seemingly ought to be supported by some conversations on a Discord server.
The concern is obvious though, not sure what's unclear about that: it's a bit pointless to have E2EE, if the adversary has full access to one of the ends anyways.
- Short of suspecting a malicious tarball, I really can't think of a reason why "publish[ing] their code on an independent third-party public code sharing platform" would be a selling point. You're getting the source code straight from the horse's mouth this way.
- Nobody took over maintenance at the time. Eventually clsid2 picked it up, and it has been maintained by him ever since.
- There's a term overload involved. In implementation terms, codec stands for coder/decoder, with "coder" referring exactly to an encoding capability: https://en.wikipedia.org/wiki/Codec
So that's a swing and a miss I'm afraid. But I'm very interested to hear what do you think a "coder" library does in this context if not encode, and why is it juxtaposed with "decoder" if not for doing the exact opposite.
- > I think there is a chronic issue of students thinking they deserve good grades, or deserve a diploma simply for showing up, in social media and I probably read that into your comment where I shouldn't have.
Yeah, that's fine, I can definitely appreciate that angle too.
As you can probably surmise, I've had quite some struggles during my college years specifically, hence my angle of concern. It used to be the other way around, I was doing very well prior to college, and would always find people's complaints to be just excuses. But then stuff happened, and I was never really the same. The rest followed.
My personal sob story aside, what I've come to find is that while yes, a lot of the things slackers say are cheap excuses or appeals to fringe edge-cases, some are surprisingly valid. For example, if this aforementioned 99% attrition rate is real, that is very very suspect. Worse still though, I'd find things that people weren't talking about, but were even more problematic. I'll have to unfortunately keep that to myself though for privacy reasons [0] [1].
Regarding grading, I find grade inflation very concerning, and I don't really see a way out. What affects me at this point though is certifications, and the same issue is kind of present there as well. I have a few colleagues who are AWS Certified xyz Engineers for example, but would stare at the AWS Management Console like a deer in the headlights, and would ask exceedingly stupid questions. The "fee extraction" practice wouldn't be too unfamiliar for the certification industry either - although that one doesn't bother me much, since I don't have to pay for these out of my own pocket, thankfully.
> If you learn the material, you pass and get a diploma. This is no more a gamble than your paycheck
I'd like to push back on this just a little bit. I'm sure it depends on where one lives, but here you either get your diploma or tough luck. There are no partial credentials. So while you can drop out (or just temporarily suspend your studies) at the end of semester, there's still stuff on the line. Not so much with a paycheck. I guess maybe a promotion is a closer analog, depending on how a given company does it (vibes vs something structured). This is further compounded by the social narrative, that if you don't get a degree then xyz, which is also not present for one's next monthly paycheck.
[0] What I guess I can mention is that I generally found the usual cycle of study season -> exam season to be very counter-productive. In general, all these "building up hype and then releasing it all at once" type situations were extremely taxing, and not for the right reasons. I think it's pretty agreeable at least that these do not result in good knowledge retention, do not inspire healthy student engagement, nor are actually necessary. Maybe this is not even a thing in better places, I don't know.
[1] I have absolutely no training in psychology or pedagogy, so take this with a mountain of salt, but I've found that people can be not just uninterested in learning, but grow downright hostile to it, often against their own self-recognized best interests. I've experienced it on myself, as well as seen it with others. It can be very difficult to snap someone out of such a state, and I have a lingering suspicion that it kind of forms a pipeline, with the lack of interest preceding it. I'm not sure that training and evaluating people in such a state results in a reasonable assessment, not for them, nor for the course they're taking.
- I don't think one applies to university expecting they're purchasing themselves a diploma, nor that they should be magically absolved of putting in effort to learn the material. What I do think is that the place they describe sounds an awful lot like people being set up for failure though, and so that begged the question as to why that might be. I should probably clarify that I wasn't particularly serious about my fraud suggestion however (was just a bit of a jab rather), as that doesn't seem to have made it through.
If teaching was so simple that you could just tell people to go RTFM then recite it from memory, I don't know why people are bothering with pedagogy at all. It'd seem that there's more to teaching and learning than the bare minimum, and that both parties are culpable. Doesn't sound like you disagree on that either.
> you're purchasing the opportunity to
We can swap out fraud for gambling if you like :) Sounds like an even closer analogy now that you mention!
Jokes aside though, isn't it a gamble? You gamble with yourself that you can [grow to] endure and succeed or drop out / something worse. The stake is the tuition, the prize is the diploma.
Now of course, tuition is per semester (here at least, dunno elsewhere), so it's reasonable to argue that the financial investment is not quite in such jeopardy as I painted it. Not sure about the emotional investment though.
Consider the Chinese Gaokao exam, especially in its infamous historical context between the 70s and 90s. The number of available seats was way lower than the number of applications [0]. The exams grueling. What do you reckon, was it the people's fault for not winning an essentially unspoken lottery? Who do you think received the blame? According to a cursory search, the individual and their families (wasn't there, cannot know) received the blame. And no, I don't think in such a tortured scheme it is the students' fault for not making the bar.
If there are fewer seats than what there is demand for, then that's overbooking, and you the test authoring / conducting authority are biased to artificially induce test failures. It is no longer a fair assessment, nor a fair dynamic. Conversely, passing is no longer an honest signal of qualification. Or rather, not passing is no longer an honest signal of unqualification. And this doesn't have to come from a single test, it can be implemented structurally too, so that you shed people along the way. Which is what I'm actually alluding to.
[0] ~4.8%, so ~95% of people failed it by design: https://en.wikipedia.org/wiki/Class_of_1977%E2%80%931978_%28...
- > Cohorts for programs with a thousand initial students had less than 10 graduates. This was the norm.
And why is this a flex exactly? Almost sounds like fraud. Get sold on how you'll be taught well and become successful. Pay. Then be sent through an experience that filters so severely, only 1% of people pass. Receive 100% of the blame when you inevitably fail. Repeat for the other 990 students. The "university thanks you for your donation" slogan doesn't sound too hot all of a sudden.
It's like some malicious compliance take on both teaching and studying. Which shouldn't even be surprising, considering the circumstances of the professors e.g. where I studied, as well as the students'.
Mind you, I was (for some classes) tested the same way. People still cheated, and grading stringency varied. People still also forgot everything shortly after wrapping up their finals on the given subjects and moved on. People also memorized questions and compiled a solutions book, and then handed them down to next year's class. Because this method does jack against that on its own. You still need to keep crafting novel questions, vary them more than just by swapping key values, etc.
- I don't recall my issues being media file or workload specific [0]. It was specifically just general frontend stuff I believe [1]. Although I should probably also mention that I don't remember much to begin with, other than my decidedly negative conclusion that made me switch players, and the overall personal narrative around that. It's been quite a few years if not a whole decade.
[0] Doesn't mean there weren't any, but then I was not doing anything special. Just watched anime, listened to music, streamed YouTube. Hardly an extraordinary workload for VLC, or indeed any media player in general.
[1] I remember them changing around the volume slider widget back and forth ad nauseam for example, and that becoming in some particular way defective that I cannot recall.
- You'll need to settle on a decoder. I personally just use my video player for this, mpc-hc.
In mpc-hc, you can framestep using CTRL+LeftArrow (steps a frame backward) or CTRL+RightArrow (steps a frame forward). This lets you select the frame you want to capture. You do not need to be on a keyframe. These keybinds are configurable and may be different on the latest version.
Then in the File menu, there's an export image option. It directly exports the frame you're currently on, to disk. Make sure to use a lossless format for comparisons (e.g. PNG).
I'm aware this can be done in other players - like mpv - as well, although there I believe no keybinds are set up for this by default, and the default export format is JPEG.
- Yes, that much was always clear. I just always thought the way these software go about finding those combinations was also standardized on a high level rather than proprietary to each implementation. It is a fairly recent development for me to realize that the various encoder options and presets are specific to the encoder, not the format (and now, that the same is true for lossless formats too).
- Drives me crazy but I'm glad to learn of it :D
Makes a lot of sense in retrospect, to the extent it bothers me I haven't figured it out myself earlier.
But who gets on the ballot is controlled by the Guardian Council. So it's like how you could get the first car Ford produced in any color, as long as that color was the color black.
Surely this is not new info to you?
> Who elects your king? Trump or Satanyahu
If you think this is a boxing or football match of some sort, please consult the forum guidelines. We're not here to beat each other (or you) down.