calf

40 ms·

571.
by calf 3y ago
Ever since the "stochastic parrots" and "super-autocomplete" criticisms of LLMs, the question is whether hallucinations are solvable in principle at all. And if hallucinations are solvable, it would of such basic and fun
572.
by calf 3y ago
Risk does not automatically remove potential conflicts of interest or vertical integration appropriateness disputes. Your argument and comparison (yes, the WeWork example is obvious self-enrichment the way you say it) makes no sense to me.
573.
by calf 3y ago
So my observation is that we could embody an AI so that it learns theory of mind-body--but then we could remove the body. This gives a theory of mindful entity that does not need a body to exist. Then the next research step could be to stud
574.
by calf 3y ago
LLMs nevertheless contain rudimentary theories, don't they? Like the Othello example demonstrating a spatial model that is emergent. LLMs are fast like calculators but it seems the optimization process that generates the parameter weig
575.
by calf 3y ago
Why is your argument from authority relevant at all here? 1. For that matter, I would not work with a boss who doesn't recognize an argument from authority as a response. 2. Furthermore a boss who lacks critical thinking skills and can
576.
by calf 3y ago
That's pathetic. I cannot respect someone and will not work under someone who functions that way. A personality cultist.
577.
by calf 3y ago
Except the article (seems to) claim that mere cognizance of self-denied thoughts and feelings is enough to make those "fall away", "force faded". There was no intermediate step of logic or problem solving involved. So, I
578.
by calf 3y ago
I couldn't follow this explanation: "Aaron could find some emotional relief because he caught sight of the anger and resentment he had so long been denying. Once he had this knowledge, the question of whether his anger was justifi
579.
by calf 3y ago
How do you reasonably distinguish this between a serious disagreement in a debate, versus an author with an axe to grind (i.e., having hidden motivations)?
580.
by calf 3y ago
That's an interesting point, are there any resources to learn about this? I have a CS background, in that we generally only cover 1st-year physics and very little theoretical math beyond linear algebra, etc.
581.
by calf 3y ago
I absolutely LOVED playing Escape Velocity in high school! AmbrosiaSW made such great shareware games, with appealing graphic design and replayability--the modular system. I think all 3 EV games are good examples of the concept of "aes
582.
by calf 3y ago
What about European regulations about genetic experimentation and GMO technology? The experiments in the broadest sense literally do not exist. They are not supposed to. I mean, the GMO regulation papers are always referring to "nascen
583.
by calf 3y ago
Apparently it does? Check this old comment from Scott Aaronson: > Rahul #97: No, because the other point is that whether a spectral gap goes to zero as the lattice size goes to infinity is the kind of thing that real condensed-matter phy
584.
by calf 3y ago
Then your critique is about LLMs specifically. But even there, can we say scientifically that LLMs cannot do math? Do we actually know that? And in my mind, that would imply LLMs cannot achieve AGI either. What do we actually know about the
585.
by calf 3y ago
But, isn't AlphaGo a solution to kind of specific mathematical problem? And that it has passed with flying colors? What I mean is, yes, neural networks are stochastic and that seems to be why they're bad at logic; on the other han
586.
by calf 3y ago
Makes me wonder again what imaginable problem it could be that a non-profit board would think it had to be kept secret.
587.
by calf 3y ago
Or, the organization has an obligation to respect freedom of academic inquiry, because the truth is what helps organizations grow, change, and mature, does it not? If the company cannot handle independent critique then its fragility is not
588.
by calf 3y ago
But is it impossible to study AI safety (or "safety") scientifically? What about all the papers like AI watermarking or AI racial bias and so for? Do you argue that academics working on those questions are doing illegitimate scien
589.
by calf 3y ago
You're correct about their error; however, Hinton views that a sufficiently scaled up autocompletion would be forced, in a loose mathematical sense, to understand things logically and analytically, because the only way approach 0 error
590.
by calf 3y ago
Tangentially I noticed that Reddit's front page has been conspicuously absent on coverage of this, I feel a twinge of pity. Maybe there are some some subreddits but I haven't bothered to look.
591.
by calf 3y ago
A Normal startup may not appeal to academics who aren't in it for the money but who want to pioneer AGI research.
592.
by calf 3y ago
This super-valuable technology would not have existed precisely because of this unstable (metastable) structure. Microsoft or Google did not create ChatGPT because internally there would have been too many rules, too many cooks, red tape, e
593.
by calf 3y ago
I think what the comment gets wrong is that even in the professor (PI) / grad student relationship, it's not defined as a purely managerial one. The one-on-one meetings between professor and graduate student are often about workin
594.
by calf 3y ago
"It's a coup" is loaded language and lets the user insinuate their position without actually explaining and justifying it.
595.
by calf 3y ago
Has anyone written a response to this paper? Their main gist is to try to define AGI empirically using only what is measurable.
596.
by calf 3y ago
The professors at the summer seminar at the Simons Insitute complained/explained (with Ilya himself present) that this research was impossible to do in university lab setting, because of the scale needed. So I would say ChatGPT exists
597.
by calf 3y ago
How many nonprofit companies are of OpenAI's size? Experience. Does Mayer have experience running a nonprofit with a charter for the good of humanity?
598.
by calf 3y ago
I wonder if there's a specific term or saying for that, maybe "projection" or "self-victimization" but not quite: when one person biasedly frames that other people were responsible for a bad thing, when it is they y
599.
by calf 3y ago
No, what is the computational complexity of the task? The computational complexity of the problem is not about how long your code is or how long you took to write it. I don't know your CS background but perhaps I do not view the terms
600.
by calf 3y ago
As a layperson, what does the special matrix transformation do? Is that the embeddings thing, or something else entirely? Something about transformer architecture I guess?

More ›