Back to News
Advertisement
Advertisement

⚡ Community Insights

Discussion Sentiment

70% Positive

Analyzed from 3952 words in the discussion.

Trending Topics

#sleep#context#model#memory#llm#term#more#weights#need#function

Discussion (139 Comments)Read Original on HackerNews

thunderbird1202 days ago
The idea of periodically stopping to write blocks of recent context into a fast-weight state is interesting, but I think it liked it better when E2E-TTT[1] did it. It's a more flexible and elegant continuous learning approach.

Essentially it goes "You know how your model can remember its training data? Well, what if you treated its recent context like more training data and updated (some of) the weights using (mostly) the same process used to train it?"

The end result is very good at remembering things but also really good at adapting to new unseen distributions.

[1]https://arxiv.org/abs/2512.23675

samsartor1 day ago
Yah I think E2E-TTT is a lot more like what people in this comments section are picturing. I can't tell that this method updates model weights at all during the "sleep" period, only the usual SSM state updated by any Mamba model after each token. They just optimized the model to use that SSM state _more_ when an eviction is about to happen.
pfannkuchen1 day ago
I wonder if we can get children to make something their life’s dream if we make the cool books about it when they are growing up? I wonder how flexible the human mind can be in convincing itself that it is fulfilling its dream?
knollimar1 day ago
This sounds like a horror novel
soulofmischief1 day ago
Each model needs to be a separate copy, or at least have those particular weights be interchangeable, for every single user.

Remember Microsoft Tay.

https://en.wikipedia.org/wiki/Tay_(chatbot)#Initial_release

thunderbird1201 day ago
Yes, since the weights being updated are a small subset of the overall total it's manageable. Just like how each separate conversation currently requires you to store a separate KV cache, you'd need to store the fast weights separately. Both KV cache and fast weight content stores have to be conversation specific, so just setting a bit of extra RAM aside for "memory" isn't really a new ask, just a different format for an old problem.
bmc75051 day ago
This topic recently came up at the FLANN workshop [1], and seems to periodically be rediscovered [2,3,4] in different contexts. While some have speculated about the biological role it plays (e.g., Pearlmutter & Houghton [5]), we still lack a conclusive theory of sleep, but the convergent evolution of this specific phenomenon across the animal kingdom and the fact that deprivation is inevitably fatal seems like an important clue.

[1]: https://flann.cs.yale.edu

[2]: https://www.cs.toronto.edu/~hinton/csc2535/readings/ws.pdf

[3]: https://arxiv.org/abs/1711.02282

[4]: https://arxiv.org/abs/2006.08381

[5]: https://mural.maynoothuniversity.ie/id/eprint/1653/1/Hamilto...

swyx2 days ago
related preprint from the letta team https://arxiv.org/abs/2504.13171

Scaling test-time compute has emerged as a key ingredient for enabling large language models (LLMs) to solve difficult problems, but comes with high latency and inference cost. We introduce sleep-time compute, which allows models to "think" offline about contexts before queries are presented: by anticipating what queries users might ask and pre-computing useful quantities, we can significantly reduce the compute requirements at test-time. To demonstrate the efficacy of our method, we create modified versions of two reasoning tasks - Stateful GSM-Symbolic and Stateful AIME. We find that sleep-time compute can reduce the amount of test-time compute needed to achieve the same accuracy by ~ 5x on Stateful GSM-Symbolic and Stateful AIME and that by scaling sleep-time compute we can further increase accuracy by up to 13% on Stateful GSM-Symbolic and 18% on Stateful AIME. Furthermore, we introduce Multi-Query GSM-Symbolic, which extends GSM-Symbolic by including multiple related queries per context. By amortizing sleep-time compute across related queries about the same context using Multi-Query GSM-Symbolic, we can decrease the average cost per query by 2.5x. We then conduct additional analysis to understand when sleep-time compute is most effective, finding the predictability of the user query to be well correlated with the efficacy of sleep-time compute. Finally, we conduct a case-study of applying sleep-time compute to a realistic agentic SWE task.

elphard1 day ago
We should let them sleep with half a brain at a time like migrating birds.
jonnyasmar1 day ago
What happened to Claude's auto-dream? I thought it was brilliant.
rahen2 days ago
That's an idea I had a few months ago: after going through a compaction once the KV cache is nearing capacity, accumulate this knowledge into a dataset to fine-tune a LoRA during offline hours.

This would create a three-layer memory system:

- Stable long-term memory (initial base weights)

- Mid-term memory built from the compactions and replay buffers

- Short-term memory (KV cache)

Sleeping would just be a fancy term for consolidating and transferring information from one memory layer to another during offline hours. Maybe that's also what the brain does while sleeping.

chermi2 days ago
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.
rahen2 days ago
I was thinking of curated replay buffers, which would act like "dreams". To prevent collapse, the offline dataset would mix the new mid-term data with a baseline of anchor data (the original training distribution) so the model doesn't drift.

Also, we wouldn't train on the whole session. A separate critic module, like a reward model, would filter the KV cache to extract the high-value information, like a garbage collector before the LoRA.

That's just an idea though. Right now most research focuses on changing the architecture itself (TITAN, HOPE...) instead.

jack_pp2 days ago
We can trust the feedback we give it based on the output it provides.
ambicapter1 day ago
What kind of feedback are you giving? What's the reward function?
DonHopkins2 days ago
It's a network of computers with GPUs, so there's no reason it can't sleep at the same time it's awake. Just a continuous "sleeping" process going on in the background, incrementally updating the model. No need for the "thinking" process to be "unconscious" while the "sleeping" process runs. Anthropomorphism confuses everything. There's no such thing as "offline hours" because the Earth is a sphere and the United States is not the center of the universe.
fc417fc8021 day ago
> the Earth is a sphere and the United States is not the center of the universe.

Felt like stating the obvious there? Greenwich being the center of everything after all.

jgreid2 days ago
Isn't this simply context pruning/optimization?
kylemaxwell2 days ago
From the abstract, it looks like it's actually doing something deeper, updating weights in part of the model?
samsartor1 day ago
The abstract and method sections only mention updating the SSM state during "sleep" (ie the same vectors that change after each token in stock Mamba) not any of the actual weight matrices. AFAICT this is just another attention compaction paper, with misleading tile? It is not very clearly written
colechristensen2 days ago
No, they're actually training weights based on context before compaction. Context is context, this is splitting the model into persistent weights and malleable ones which are periodically updated.
delis-thumbs-7e2 days ago
Wouldn’t that be extremely computationaly expensive considering how resource incentive training is?
colechristensen2 days ago
No, training a state of the art model involves training on the order of 10 trillion tokens.

We're talking about a step that updates weights based on say between 10k and 1M tokens.

energy1231 day ago
Would be a big deal if you don't have to care about quadratic attention cost. Some workflows become a lot cheaper.
hmokiguess1 day ago
This could be a solution in search of a problem, I would be careful with overfitting.
mos765817about 22 hours ago
wasn't this what Google did long ago? https://openreview.net/forum?id=iiZy6xyVVE
Advertisement
gt01 day ago
This seems as much like "sleep" as when a laptop "sleeps".
m0unta1ntube1 day ago
why not just design the LLM like an OS?
scotty792 days ago
Context -> Lora would be soooo cool.
micromacrofoot2 days ago
To reach a more brain-like behavior LLMs need to integrate your inputs into their model dynamically, essentially retraining real-time based on the most salient input. Human brains do this selectively all the time and it's part of our plasticity.

Biologically humans do similar compression, so introducing a similar concept to an LLM also feels reasonable. Hardware isn't fast/cheap enough to do this on an ongoing basis, similar to how it's too expensive for our brains to do this while we're moving through the world.

All we have now most of the time in LLMs is "working memory" we're missing a lot of the functionality that allows for episodic memory and selective plasticity.

The more you read about how human brains work, the more you realize that we may have figured out a piece with LLMs, but it's certainly nothing approaching AGI. People insisting so are blowing smoke for investor hype or don't understand a big piece of the concepts involved.

logicchains1 day ago
>To reach a more brain-like behavior LLMs need to integrate your inputs into their model dynamically, essentially retraining real-time based on the most salient input.

That's already possible with LLMs. The challenge is that 1. it would allow permanently jail-breaking models and 2. there'd be no way for them to efficiently transfer what they'd learned to a new model generation.

micromacrofoot1 day ago
Oh do you have a source? I haven't seen it done in real-time.

Coincidentally the human brain is also jailbroken and nontransferable

semiinfinitely1 day ago
academic clickbait
m3kw91 day ago
sleep aka processing the data differently.
pcrh2 days ago
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?

djeastm2 days ago
They provide an explanation for using the term "sleep":

> In animals, the transfer from short-term memory to long-term memory is thought to be supported by hippocampal replay [33], especially during sleep [41]; in this phase, short-term hippocampal memories are reactivated and consolidated into cortical synaptic weights. Sleep makes animals unable to respond to external stimuli, suggesting that it must provide enough cognitive benefit to justify this cost [41]. Inspired by these biological processes, we propose a method for transferring context-window memory into persistent weights. When the model’s context window becomes full during inference, the model enters a “sleep” in which it performs multiple forward passes over the accumulated context and recursively updates its fast weights via a learned local rule. As in animal sleep, the model receives no external input tokens during this phase. After consolidation, the context window is cleared, and the model resumes operation with updated fast weights. During training, the model is optimized end-to-end by backpropagating through the entire process to maximize task performance after sleep.

pcrh2 days ago
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?

bayarearefugee2 days ago
> Does their LLM "die" if it can't perform the function described?

If you don't periodically clean the context, an LLM effectively goes insane in terms of outputs.

If the LLM were fully controlling a physical system (like a robot body) that contained it the resulting insanity of an ever-growing, never cleaned context would likely result in some sort of death-like event.

adastra222 days ago
There is a lot that is known about sleep. We don't know everything and there are large gaps in our knowledge, but there is also a lot that we do know. And this research explicitly tried to emulate the things we know that sleep does do. Calling it "sleep" is warranted, imho.
ben_w2 days ago
> If an animal can't sleep it will eventually die.

Very few animals fail to eventually die even with as much sleep as they want.

But before death, there is a loss of cognitive function from sleep deprivation, and we observe this too with AI whose context windows get too full.

While we don't know very much about sleep, my understanding is that we do have a long list of things that we do during it, we just don't really understand if sleep is necessary for each of them or simply a convenient opportunity for it.

There's lots of things biology does in response to easy-to-detect proxy signals instead of the real thing they care about: Our sensation of needing to breathe more is based on too much carbonic acid in our blood, not lack of oxygen, which is why in general nobody is allowed in an elevator with a liquid nitrogen dewar; Our natural distaste for incest is based on who we grew up with, not our actual DNA; Get too cold and some people suddenly feel warm and want to (and some do) take all their clothes off even though that would just make them hypothermic even faster.

Being asleep may trigger the things we need to get done, but that doesn't mean sleep is *fundamentally* necessary for the things we need to get done. It could be just that it happens to be the way our biochemistry is wired, and we may find some other way to trigger those things.

The quotation given by djeastm would by my guess for what a dream is, and why we have them. But we don't spend all our time asleep, dreaming. And I'd be the first to say that my guess isn't worth much, as I'm not a brain scientist.

nextaccountic1 day ago
> The function of sleep in animals is largely obscure.

This just obscures the conversation IMO. We know a great deal of functions that sleeping performs. We don't know everything. For some reason this prevents us for using this word in a computing context? What do you think about sleep(1) in Unix?

Also..

> Does their LLM "die" if it can't perform the function described?

what "death" means in the context of a computer program?

palmotea2 days ago
> The function of sleep in animals is largely obscure.

Also, there's different kinds/stages of sleep, which probably perform different functions.

For instance, REM may do something like the GP describes, consolidating memories and processing learning. Deep sleep may do something else (I vaguely recall some stage of sleep is used by neurons to clear certain waste products).

Windchaser2 days ago
I don't think it's necessarily correct to think of sleep in terms of "it is necessary for animals or they will die". It might be more useful to think of it as "it was so useful that animals who slept outcompeted all the animals who didn't".

Meaning: it might just provide a big advantage.

I don't want to overextend and assume that any advantage extends to LLMs. That rest-and-recuperate advantage might also extend to LLM-based AIs. Or maybe not, and the rest-and-recuperate is mainly useful for biology-based organisms. But there is some logic to it.

> The function of sleep in animals is largely obscure.

In my understanding, it's well-understood that sleep is used to consolidate and store long-term memories (amongst other functions, like cell and muscle repair). They've found this memory-consolidation-during-sleep even in relatively simple animals like bees.

crabbone1 day ago
I think sleep serves multiple functions. For example, anyone who works out in any-what systematic way knows that sleep is essential for muscle grow. You can't skip on sleep if you want to get fitter. And this probably has very little to do with the more sophisticated functionality of the brain, rather it allows for some process in muscle tissue to happen.

So, whether the LLM "dies" in any sense may or may not be important for what "sleep" is defined to be in this article. It's quite possible that sleep also affects endocrine system in animals or hormones etc... and that's what's causing death, not necessarily anything to do with how brain functions.

exe341 day ago
LLMs also don't mate, but we can still talk about how they remember or forget things. The meaning of words change. Some of those changes are useful.
naasking1 day ago
> Does their LLM "die" if it can't perform the function described?

It dies in terms of usefulness if it can't stay up to date with new knowledge. That is, it will no longer be used and thus effectively die off.

sillysaurusx2 days ago
> 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 other words, it could be true that “among vertebrates, only rats die of sleep deprivation.”

So “if an animal can’t sleep, it will eventually die” is actually quite hard to prove, and depending on how you look at it, somewhat easy to disprove by the fact that rats and fruit flies were so difficult to kill from sleep depravation alone.

Personally I’m skeptical of the rat study too. Claude amends this:

> What they did not establish: the mechanism. On autopsy, “no anatomical cause of death was identified.” The rats showed weight loss despite eating more, body temperature problems, and skin lesions, but nothing that pointed to a clean cause. So no, they could not say a rat “died from sleep deprivation alone” in the sense of identifying what sleep loss did to the body to kill it. They showed a strong association under tight controls, not a proven causal pathway.

libria1 day ago
Is a volcano described as dormant (dormire, literally sleep) also inaccurate and deeply problematic? BTW, it's not anthropomorphized as sleep has existed long before humans.

"Sleep" is just used in their context to describe a non-interactive mode and they didn't lean heavily into zoomorphic - I think you mean - parallels.

You're grinding an axe on a single term. What is your broader hangup with them using the term "sleep"?

> Does their LLM "die" if it can't perform the function described?

We're reaching an age where LMGTFY should now be Let Me LLM That For You. Have you tried asking an LLM this question about the article? I believe it answers it very well.

Kaliboy1 day ago
This is very interesting to me, I've been sleeping a lot lately.

I'm autistic and just went through massive changes that basically keep locking up my brain and I freeze.

I learned that sleeping helps, even if just 20 minutes. It helps that I can fall asleep "while awake". It's as if I relinquish control of responding to stimuli, which instantly brings so much rest to my mind. It is odd trying to move a limb and the brain basically responds with noop. But it works.

Afterwards I can generally make a decision and perform it.

So in a sense it seems similar to what you describe the model would have to do. I forget short term concerns that overwhelm and refocus on the long term goal.

order-matters2 days ago
but isnt sleep an already defined technical term for significantly reducing power consumption while preserving its state until woken up?

i feel like its confusing to reuse the word for a process that aims to deliberately change state of the machine / process

raincole2 days ago
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!
not_a_bot_4sho1 day ago
Some of them were straight up psychopaths too, as evidenced by `kill()` !
prerok1 day ago
Indeed and using SIGKILL is really cruel. At least with SIGTERM the process can say its goodbyes. /j
famouswaffles2 days ago
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.

pfdietz1 day ago
We shouldn't anthropomorphize LLMs. They hate it when you do that.
forshaper1 day ago
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...
gabriela_c1 day ago
Feels like we're having a computer world Jane Goodall moment.
DonHopkins2 days ago
Is it "Anthropicmorphization" when Claud treats human beings like LLMs?
sillysaurusx2 days ago
Interesting question. Is there an actual term for that? It’s like inverse anthropomorphization, but not quite.
CuriouslyC1 day ago
Saying something needs sleep isn't anthropomorphizing, since pretty much all complex living organisms need sleep.

Also, even when something is "specific" to humans, it might not be anthropomorphizing to observe it in something else, it could just be an emergent pattern of high intelligence.

gchamonlive1 day ago
Just like LLM sleep has nothing to do with animal sleep, the neuron in a neural network has nothing to do with an actual neuron, and nobody should pretend they do.

I agree we need to be mindful of our metaphores, but they do help both with inspiration for developing techniques as well as for naming things. The onus of keeping bias in check when using metaphores is on the reader, authors can't really do that for you. However once bias is in check you can have a very productive debate in terms of these namings given that everyone is aware of their ontology.

ajs19982 days ago
This is the struggle of naming papers. You could stretch definitions and make your own sexy headline or you could be precise and fewer people will read it.
cush1 day ago
I think it's interesting that folks are suddenly taking issue with "anthropomorphizing" language used in AI as if we haven't been doing this since the earliest days of computing (see "memory", "child", "parent", etc). It helps folks understand things at the correct level without needing domain knowledge
ComplexSystems1 day ago
That's because the purpose of this article is not to have an objective debate over their abilities at all. Most interesting research in this field isn't. Instead, it's to present a new technique to improve LLM performance, which is much more interesting than (once again) rehashing the philosophy of LLM personhood.
Dusseldorf1 day ago
Was it anthropomorphizing computers when they named "memory"? Seems to me like it's more analogizing for the sake of easy understanding. Sure, it's not literally the same exact mechanism, but it's certainly modeled after the biological concept.
lxgr2 days ago
If it works, it's called bionics, not anthropomorphization ;)
skeledrew1 day ago
How do you concisely describe a low power state of an entity that processes, whereby when in that state it has little to no reaction to input and it may or may not be performing tasks in that state, for a mixed education audience?

Also keep in mind that most if not all devices with a chip have had a function called "sleep" for many years, without this argument.

burnte1 day ago
> 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.

eithed2 days ago
I assume compacting is the sleep here; so, yes
motoxpro1 day ago
One of the most common functions in programming is sleep(ms). There is wake, heartbeat, handshake, orphan, listen, starve, parent/child, etc.

This is not anything new, its just a word that fits the function.

FarmerPotato1 day ago
>Does a motor vehicle get "sleep" when it is serviced? When I reboot a computer, is that equivalent to a nap?

Do androids dream of electric sheep?

jonnyasmar1 day ago
When the goal of that function is to think (a notoriously human behavior), it's perfectly understandable to anthropomorphize it.
madibo31561 day ago
I find this annoying too. "Sleep" is okay, but the quippy headlines ("need sleep"—short, snappy and vague) infiltrating journals bother me. I've seen it well before LLMs, but as an example, there is a long list of title snowclones of the famous attention paper: https://github.com/vinayprabhu/X-is-all-you-need.
tom_2 days ago
throw3108221 day ago
First, this is not a "debate over the abilities" of LLMs. It's a proposed method to improve their performance, and the authors are free to call it however they think it makes sense.

Second, explicitly avoiding things that sound like anthropomorphisation is equally not helpful- why avoid a metaphor that works?

Third, it's really a pity that this pointless nitpicking is dominating the thread.

pcrh1 day ago
>this pointless nitpicking is dominating the thread.

What is dominating the thread are claims that the LLM operation in question is analogous to the function of sleep in humans. It obviously is not.

The anthropomorphization of LLMs has reached ridiculous proportions. Applying the same standards as used in this field to others would result in claims that laundry machines "hallucinated" that they had had sufficient water when they failed due to the faucet being turned off.

genxy1 day ago
Please re-read up to the end of page 2 and then re-ask this question.
zeckalpha1 day ago
Yet this is how "thinking" models got started.
simonw2 days ago
> When I reboot a computer, is that equivalent to a nap?

I mean, you do put your computer into "sleep" mode and then "wake" it.

Analogies are useful. I think we need to learn how to continue to benefit from them despite the risk of anthropomorphication.

cowlby2 days ago
The analogy is helpful, but yes we should be able to “intelligently design” something better than sleep analogues since we’re not constrained by evolution like in humans.
SR2Z2 days ago
Evolution constrains the evolution of human beings, but it's also excellent at discovering elegant designs that work very reliably at a low cost.

Maybe someday we'll understand the way our minds work well enough to design from first principles but until then we've only got one template for how a thinking machine should look.

lxgr2 days ago
We are however constrained by the complexity of any purported solution. That's the bitter lesson, in a nutshell.

At the very least, we know that sleep and dreaming do exist in biological brains. (Doesn't mean any of it is applicable to artificial neural nets, doesn't mean it'll work for our specific architectures etc. etc., but at least the idea requires fewer assumptions than a completely untested novel theory.)

reaperducer1 day ago
Does a motor vehicle get "sleep" when it is serviced?

One of the mayors of New York in the 80's (Koch?) famously doubled the city's bus fleet for zero cost by running them 24 hours, instead of letting them rest at the end of their shifts, as was the previous policy.

wat100002 days ago
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.

aaroninsf2 days ago
Very much agree that while it is is useful in description of motivation and inspiration,

it is very non-helpful—or worse—to use this language, this way.

One might as well say "need neural plasticity" which is as much an analogy and equally misleading and counterproductive in shaping the right model of the system.

One might even call this pernicious, what it encourages is already a social problem; and it doesn't aid understanding, it confounds it.

colechristensen2 days ago
>we study a sleep-like consolidation mechanism in which a model periodically converts recent context into persistent fast weights before clearing its key-value cache

There is a strong, non-trivial connection here between what your brain does in sleep and what they are studying.

You wouldn't object to referring to robot eyes or robot legs.

verisimi2 days ago
... and anyway, maybe it was hungry? Or getting the sniffles?
IAmGraydon1 day ago
The entire industry is so desperate to anthropomorphize. What the paper describes is an offline recurrent consolidation phase: the model runs multiple forward passes over recently accumulated context, updates persistent fast weights in SSM blocks, then clears the KV cache before continuing. It has absolutely nothing to do with sleeping, but I believe the authors had a goal in mind when creating this title, and it was for journalists to pick it up and run with it, further inflating the AI-is-just-like-us hype bubble.
genxy1 day ago
It is a descriptive analogy, get over yourself.
IAmGraydon1 day ago
An intelligent reply from an obviously intelligent guy!

A more appropriate title would have been something like "Offline Recurrent Memory Consolidation for Long-Context Language Models". This is supposed to be a research paper, not a story book. The title should give context to other researchers, and not be clearly engineered for clicks. If you don't think so, that's your prerogative, but you're objectively wrong.

genxy1 day ago
You write the paper, you write the title. So much anger over a title, you are graydon, make this about yourself.
hansmayer1 day ago
Sweet Jesus, so not only are they performing qualitatively worse than humans, too expensive for any serious work, but now they also "need" to sleep? What's next - unionisation so they can enjoy 8 hours of culture too?
victorkulla1 day ago
No they do not. I'm sure that if you presented the same argument about, I don't know?, your car's CPU with built in AI; then this would be a whole different discussion entirely.
Advertisement
danielrmay1 day ago
The "sleep" thing gives me the creeps so in my head I'm just going to think of it as the difference between "response time retrieval" and "background consolidation".

I do think it points at something bigger than just attention architecture: "memory" isn't just storage, and merely longer context isn't the same thing as having a better understanding of the source data.

I'm looking at this through the "personal AI" lens, where I think the missing "memory" layer seems to be consolidation & prioritization. It's not enough to just pattern match and grab the right emails, notes, etc, stuff them into the context window & hope, but instead it's useful to consider offline processing and turn events into durable state: clusters of observed data becomes episodes, assumptions, contradictions and power confidence for suggestions.

That also pushes up the need for provenance & inspectability. It's going to be interesting to see what kind of memory consolidation strategies are required for each domain use case.

sonink1 day ago
I think you are missing the most important part - forgetting. The missing "memory" layers is consolidation, prioritization AND forgetting (what is not important).

Also not too sure about provenance and inspectability - it is part of memory. If the source is deemed 'important' it will survive forgetting. If not, then maybe not. And its ok. I am sure you dont know the exact source who told you that the capital of France is Paris. You forgot, and its no big deal.