Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

A sleep-like consolidation mechanism for LLMs

arxiv.org

Post reply on HN
Print view View original

51–60 of 155 posts

Re: A sleep-like consolidation mechanism for LLMs

#51

Post by wat10000 »

pcrh wrote: post #4

I can't pretend to understand how LLMs work, but I can be sure that anthropomorphizing their functions is not helpful to an objective debate over their abilities. Does a motor vehicle get "sleep" when it is serviced? When I reboot a computer, is that equivalent to a nap?

Just from the title, I’m assuming it refers to a period of downtime used to perform some sort of maintenance on the knowledge held by the system.

Clicking through, that’s exactly what it is. Seems like “sleep” is an excellent term to use here.

Re: A sleep-like consolidation mechanism for LLMs

#52

Post by forshaper »

famouswaffles wrote: post #12
pcrh wrote: post #4

I can't pretend to understand how LLMs work, but I can be sure that anthropomorphizing their functions is not helpful to an objective debate over their abilities. Does a motor vehicle get "sleep" when it is serviced? When I reboot a computer, is that equivalent to a nap?

Anthropomorphization is not inherently wrong, and in some instances, it actually lets you reason better about about complex behavior than whatever convoluted (and often wrong, especially in the case of giant neural networks) mechanistic description one might conjure. Here the analogy isn't without reason.

Wason Selection task performance improvements based on social framing suggest that it's easier for us to think about problems when some anthropomorphization is going on. https://www.cep.ucsb.edu/wp-content/uploads/2023/05/Cogadapt...

Re: A sleep-like consolidation mechanism for LLMs

#53

Post by ambicapter »

jack_pp wrote: post #36
chermi wrote: post #28

Earlier quoted context omitted.

Wouldn't that just accelerate collapse? How much do you trust the outputs of the llm to provide trustworthy and valuable new information? I mean I understand distillation works. But that's much more structured and thoughtful than my sessions at least.

We can trust the feedback we give it based on the output it provides.

What kind of feedback are you giving? What's the reward function?

Re: A sleep-like consolidation mechanism for LLMs

#54

Post by burnte »

pcrh wrote: post #4

I can't pretend to understand how LLMs work, but I can be sure that anthropomorphizing their functions is not helpful to an objective debate over their abilities. Does a motor vehicle get "sleep" when it is serviced? When I reboot a computer, is that equivalent to a nap?

> Does a motor vehicle get "sleep" when it is serviced?

That's more like a doctor visit and a workout. The sleep will be the part of the duty cycle when it's not operating.

> When I reboot a computer, is that equivalent to a nap?

Yes, it wakes up completely refreshed and in good working order, usually, and if there's still a problem you know you need a technician.

Re: A sleep-like consolidation mechanism for LLMs

#55

Post by sillysaurusx »

ambicapter wrote: post #49
sillysaurusx wrote: post #43

Earlier quoted context omitted.

> If an animal can't sleep it will eventually die. That turns out to be un-settled science. No human has ever died from lack of sleep. People point to “fatal familial insomnia” as a counterexample. But they die to the disease, not the lack of sleep. In a series of controlled experiments, rats and fruit flies did die from lack of sleep. But no one has yet proven that it holds true for vertebrates except for rats. In o…

So? You don't need a proven causal pathway to state that a glass heads towards the ground every time you brush it off a table.

Scientifically you do, otherwise you can’t claim that lack of sleep was the cause of death. It could be an artifact of how the experiment was run, or any number of other factors.

It’s not a small quibble to point out that the central argument (“animals need sleep or they’ll die”) may be mistaken.

Re: A sleep-like consolidation mechanism for LLMs

#56

Post by dijit »

sillysaurusx wrote: post #43
pcrh wrote: post #20

Earlier quoted context omitted.

The function of sleep in animals is largely obscure. One thing we do know for certain is that it is necessary, it is needed in "dumb" animals as well as in you and I. If an animal can't sleep it will eventually die. I don't think that applies to the activity described in the OP. Does their LLM "die" if it can't perform the function described?

> If an animal can't sleep it will eventually die. That turns out to be un-settled science. No human has ever died from lack of sleep. People point to “fatal familial insomnia” as a counterexample. But they die to the disease, not the lack of sleep. In a series of controlled experiments, rats and fruit flies did die from lack of sleep. But no one has yet proven that it holds true for vertebrates except for rats. In o…

> No human has ever died from lack of sleep.

As far as I understand it, there is a disease that destroys your brain's ability to produce sleep. Once you have it, you suffer total, progressive insomnia and die within roughly 6–18 months. Scientists debate whether it's the underlying brain damage or the sleeplessness itself that causes death, but the two are inseparable in practice, and sleep deprivation is considered the leading candidate.

Separately, the longest anyone has stayed awake under controlled conditions was 11 days, which produced severe cognitive impairment, paranoia, and hallucinations; suggesting the body deteriorates rapidly without sleep.

It's probably not wise to state your original claim as established fact.

Re: A sleep-like consolidation mechanism for LLMs

#57

Post by burnte »

sillysaurusx wrote: post #43
pcrh wrote: post #20

Earlier quoted context omitted.

The function of sleep in animals is largely obscure. One thing we do know for certain is that it is necessary, it is needed in "dumb" animals as well as in you and I. If an animal can't sleep it will eventually die. I don't think that applies to the activity described in the OP. Does their LLM "die" if it can't perform the function described?

> If an animal can't sleep it will eventually die. That turns out to be un-settled science. No human has ever died from lack of sleep. People point to “fatal familial insomnia” as a counterexample. But they die to the disease, not the lack of sleep. In a series of controlled experiments, rats and fruit flies did die from lack of sleep. But no one has yet proven that it holds true for vertebrates except for rats. In o…

HIV doesn't kill you, but it creates circumstances where other things will. Sleep is the same. You may not die from lack of sleep, but you die from the things it can cause. Effectively there's no difference.

Re: A sleep-like consolidation mechanism for LLMs

#58

Post by gabriela_c »

famouswaffles wrote: post #12
pcrh wrote: post #4

I can't pretend to understand how LLMs work, but I can be sure that anthropomorphizing their functions is not helpful to an objective debate over their abilities. Does a motor vehicle get "sleep" when it is serviced? When I reboot a computer, is that equivalent to a nap?

Anthropomorphization is not inherently wrong, and in some instances, it actually lets you reason better about about complex behavior than whatever convoluted (and often wrong, especially in the case of giant neural networks) mechanistic description one might conjure. Here the analogy isn't without reason.

Feels like we're having a computer world Jane Goodall moment.

Re: A sleep-like consolidation mechanism for LLMs

#59

Post by sillysaurusx »

dijit wrote: post #56
sillysaurusx wrote: post #43

Earlier quoted context omitted.

> If an animal can't sleep it will eventually die. That turns out to be un-settled science. No human has ever died from lack of sleep. People point to “fatal familial insomnia” as a counterexample. But they die to the disease, not the lack of sleep. In a series of controlled experiments, rats and fruit flies did die from lack of sleep. But no one has yet proven that it holds true for vertebrates except for rats. In o…

> No human has ever died from lack of sleep. As far as I understand it, there is a disease that destroys your brain's ability to produce sleep. Once you have it, you suffer total, progressive insomnia and die within roughly 6–18 months. Scientists debate whether it's the underlying brain damage or the sleeplessness itself that causes death, but the two are inseparable in practice, and sleep deprivation is considered…

My second paragraph addresses that:

> People point to “fatal familial insomnia” as a counterexample. But they die to the disease, not the lack of sleep.

It’s a prion disease. It’s established fact that they don’t die from the lack of sleep.

Re: A sleep-like consolidation mechanism for LLMs

#60

Post by not_a_bot_4sho »

raincole wrote: post #16
pcrh wrote: post #4

I can't pretend to understand how LLMs work, but I can be sure that anthropomorphizing their functions is not helpful to an objective debate over their abilities. Does a motor vehicle get "sleep" when it is serviced? When I reboot a computer, is that equivalent to a nap?

This is why I object to sleep() from unistd.h. What an anthropomorphizing notion. Didn't early unix programmers understand that a computer isn't a living creature and therefore isn't capable of sleep? They must have been really stupid!

Some of them were straight up psychopaths too, as evidenced by `kill()` !
Post reply on HN