[HN Gopher] Unpredictable abilities emerging from large AI models
___________________________________________________________________
Unpredictable abilities emerging from large AI models
Author : nsoonhui
Score : 165 points
Date : 2023-03-17 10:04 UTC (12 hours ago)
(HTM) web link (www.quantamagazine.org)
(TXT) w3m dump (www.quantamagazine.org)
| sublinear wrote:
| From the review comments...
|
| > Key Weakness: The paper largely focuses on showing how much
| emergence occurs in a "sudden" manner, bringing reports from
| previous work. It relies on the "magic" of emergence, rather than
| providing new insights on why this is happening and when it
| happens/does not happen.
|
| > Requested change: More fundamental evidence on the claim
| "further scaling will likely endow even-larger language models
| with new emergent abilities" with more concrete discussion _(with
| possibly evidence)_ on how those new emergent abilities would
| look like and how further scaling will be possibly in a
| approachable way.
| mariodiana wrote:
| What happens when AI models ingest the AI-generated content that
| will soon flood the web?
| aimbivalent wrote:
| I write this as a sceptic who always leaned towards the symbolic
| side of this debate. First seeing Open AI inventing tactics in
| Dota and now using GPT and other recent AI product has changed my
| mind. The rate of progress in the field is just staggering and
| seems to be still accelerating.
|
| Overall, I'd say we are running into too many "unkowns". We don't
| know our own cognition works which makes all arguments along the
| lines of "LLMs are just .." moot. We don't even really know how
| GPT (and others) do what they do and so all arguments along the
| lines of "LLMs won't be able to ... " are similarly moot.
|
| I'm not really taking a position except to say that most
| arguments I see most people making are moot. So, nobody, and me
| least of all, has a clue about how all of this might shake out.
| RC_ITR wrote:
| >We don't know our own cognition works which makes all
| arguments along the lines of "LLMs are just .."
|
| Sure, but there are very binary tests we can do to understand
| the first principles of what LLMs are vs. what they are not.
|
| Ask an LLM to play tic-tac-toe and it does great. Ask it to
| play tic-tac-toe on a 100x100 board, it get's confused.
|
| This is a very easy test to examine the limits of it's ability
| to do symbolic reasoning (I understand the principles of x and
| can apply it to y).
|
| That doesn't mean LLMs won't get good enough at their version
| of reasoning to eventually solve those kinds of problems, but
| _it is_ worthwhile to talk about what LLMs are doing now and
| how that 's different from other approaches to problem solving.
|
| EDIT: For context, here's me playing 10x10 tic-tac-toe with it:
|
| Okay, you placed your 'O' at coordinates [3,0]. Here's the
| current board: 0 1 2 3 4 5 6 7 8 9
|
| 0 O _ _ _ _ _ _ _ _ X
|
| 1 O _ _ X _ _ _ _ _ _
|
| 2 O _ X _ _ _ _ _ _ _
|
| 3 O _ _ X _ _ _ _ _ _
|
| 4 O _ _ _ X _ _ _ _ _
|
| 5 _ _ _ _ X _ _ _ _ _
|
| 6 _ _ _ _ _ _ _ _ _ _
|
| 7 _ _ _ _ _ _ _ _ _ _
|
| 8 O _ _ _ _ _ _ _ _ _
|
| 9 O _ _ _ _ _ _ _ _ X
|
| You have managed to get four 'O's in a row, so you win the
| game! Congratulations!
|
| ^ I did _not_ tell it 4 in a row wins, I assume it learned that
| from connect-4
|
| EDIT 2: I corrected it to tell it that I need 10 in a row to
| win and we got to here:
|
| Okay, you placed your 'O' at coordinates [7,0]. Here's the
| current board: 0 1 2 3 4 5 6 7 8 9
|
| 0 O _ _ _ _ _ _ _ _ X
|
| 1 O _ _ X _ _ X _ _ _
|
| 2 O _ X _ _ _ _ _ _ _
|
| 3 O _ _ X _ _ _ _ _ _
|
| 4 O _ _ _ X _ _ _ _ _
|
| 5 O _ _ _ X _ _ _ _ _
|
| 6 O _ _ _ _ _ _ _ _ _
|
| 7 O _ X _ _ _ _ _ _ _
|
| 8 O _ _ _ _ _ _ _ _ _
|
| 9 O _ _ _ _ _ _ _ _ X
|
| You have managed to get seven 'O's in a row, but you still need
| three more to win. It's my turn again. I'll place an 'X' at
| coordinates [6,9].
| aimbivalent wrote:
| hey that's pretty cool. Yeah, for sure it's worthwhile
| talking about what they are doing, I agree. I guess I just
| wish people would update their mental model to the new
| reality a bit quicker :)
| wslh wrote:
| How does it work regarding queries in natural language? I
| mean, thinking on translating a natural language question to
| an SQL query in complex scenarios.
| tel wrote:
| I've been asking GPT-4 to design whole systems for me off
| of sparse natural language specifications. It gives
| reasonable designs, I read and critique, it updates and
| modifies. I regularly run into limitations, sure, but it
| will likely blow you away with its capability to convert
| natural language questions to SQL---given adequate specific
| context about your problem.
| novaRom wrote:
| Basic statistical relations in observable world (on Human-Level
| scale) should somehow be reflected in human language and in
| mathematics. Deeper models tend to learn those relations, and
| produce good looking connections which fit in the realm of
| possible "thoughts" perfectly.
| anaverageindian wrote:
| Google engineers will go on and talk about what they discovered
| in LLMs in 2020 rather than just releasing the product.
| _facepalm_
| interdrift wrote:
| In the past few days I've been shellshocked by GPT-4. I'm still
| and believe I will be for the near future.
| mark_l_watson wrote:
| Nice write up! I have been using classic back-prop neural
| networks since the 1980s, and deep learning for the last 8 years.
| This tech feels like a rocket ship that is accelerating
| exponentially!
|
| I am in my 70s and I don't work much anymore. That said, I find
| myself spending many hours in a typical day doing what I call
| "gentleman scientist" activities around Large Language Models.
|
| I was walking this morning with a non-tech friend and I was
| trying to impart to him my feeling that all knowledge work, and a
| lot of mundane work is getting disrupted. Even though I have many
| tech-friends who are "Large Language Model doubters", my friend
| got it this morning. Emotionally and psychologically I think that
| some tech and other highly paid knowledge workers just can't
| accept the sea-change that we are living through.
|
| For myself, I am spending a very large amount of time
| experimenting with the OpenAI APIs, LangChain, and Llama-Index -
| and I am enjoying myself tremendously.
| anon7725 wrote:
| > Emotionally and psychologically I think that some tech and
| other highly paid knowledge workers just can't accept the sea-
| change that we are living through.
|
| This is a good way to frame it. As a mid-career person, I'm
| trying to figure out how to respond to these developments in my
| own career.
|
| I think there's a good chance that software development as a
| career looks very different in 10 years in terms of the day to
| day work and opportunities for employment.
|
| 25 years ago in university, I did roughly the same stuff as I
| do today. Languages are higher on the abstraction ladder,
| networks and hardware are faster, etc, but stuff is still
| roughly the same. I can see a real discontinuity in the work on
| the horizon.
| roflyear wrote:
| I think it'll end at using chatgpt as a great tool. And that
| seems fine. I think much of the anxiety here is in the
| unknown.
|
| You have to learn your IDE, but it doesn't make programming
| harder. It makes it more fun actually.
|
| Chatgpt is going to require that devs are even better IMO
| because you have to catch its errors and know how to work
| with it.
| UncleOxidant wrote:
| Yes, this stuff is fun to work with and use as an aid to
| programming. But I think what's dawning on a lot of us is
| that it seems to be able to facilitate a large increase
| productivity. And given that large productivity increase
| will industry need as many software developers (and other
| knowledge workers)?
| bcrosby95 wrote:
| Your opinion on this probably depends on what your job
| currently looks like. Everywhere I've worked there's been
| significantly more stuff we wanted to do than we could
| actually do. Speeding that up looks awesome to me.
|
| But if you work at BigCo and mostly sit on your hands
| already, then, yeah. I don't know.
| fnordpiglet wrote:
| Every BigCo I've worked at (and I've worked at places
| that put the B in Big) people aren't sitting on their
| hands because there isn't way more to do than they can
| feasibly do. It's usually because they're fundamentally
| illsuited for the job and sitting on their hands is
| better than being homeless, or the people who are
| illsuited for their jobs feel anxious about being
| homeless so force everyone into meetings and team
| building and off sites and retrospectives and sprint
| planning and burn down chart analysis and whatever else
| they can conceive of that seems productive but doesn't
| change production in any way. To that extent at least an
| AI would be left alone to write actual code, and those
| folks can have their endless meetings and still justify
| their job the same way they ever did. No AI will ever
| host a breakout session at an offsite to discuss strategy
| for getting people to work in the horrific hotel seating
| dehumanized office, they're too capable.
| coldtea wrote:
| > _Everywhere I 've worked there's been significantly
| more stuff we wanted to do than we could actually do.
| Speeding that up looks awesome to me._
|
| Does it? Slowing our breakneck anti-human business pace
| and stopping mindless and environmentally and culturally
| harmful consumption and production of trinkets is surely
| better!
| simonw wrote:
| I can already feel it dramatically increasing my personal
| productivity. I expect it's going to eventually lead to a
| 2x or 3x time productivity increase for software
| engineering generally (if not more).
|
| Will that result in less demand for software engineers? I
| doubt it. I think it will lead to companies and
| organizations doing way more software engineering.
|
| I was going to use the old "ATMs resulted in more jobs
| for bank tellers example", but apparently that particular
| example is flawed:
| https://www.vox.com/2017/5/8/15584268/eric-schmidt-
| alphabet-...
| ip26 wrote:
| Somewhat similar boat; I just hope it winds up helping me do
| my job better, instead of me trying to teach it how to do my
| job. Setting aside professional anxieties, after using
| chatGPT for a bit I quickly realized I am completely
| uninterested in prompt engineering. But as a tool, like with
| an IDE, it could be a big leap forward.
| bentcorner wrote:
| I literally just finished writing some documentation and
| had Copilot running, sometimes it'd autocomplete something
| with garbage (and I would just type what I wanted anyway),
| more than once it autocompleted with several lines of
| exactly what I wanted to write. This is way better than the
| phone-type autocomplete I've seen in gmail where it just
| guesses the next word.
| fnordpiglet wrote:
| I ducking agree
| FpUser wrote:
| I have started using it for a few days. Basically asking to
| write and modify pieces of code, for now small ones where I
| can easily spot a bullshit. I am very much looking forward
| for it to improve to the point where I can spend more time
| thinking about business domain rather then how to translate
| it to efficient code.
| olalonde wrote:
| Indeed, it's fascinating to witness a sizable segment of the HN
| community and distinguished intellectuals like Chomsky
| displaying absolute denial. I've started bookmarking the
| comments so I can look back at them in 5 years and have a good
| laugh. Some from a few months back are already aging badly[0].
|
| [0] https://news.ycombinator.com/item?id=34197033
| WC3w6pXxgGd wrote:
| Has Chomsky been right about anything?
| HyperSane wrote:
| About foreign policy? No.
| fnordpiglet wrote:
| Given how emphatic he is about everything, the answer is
| clearly no.
| Izkata wrote:
| My first introduction to him was when one uncontacted tribe
| or anothers' language refuted something he thought was
| fundamental to humans.
|
| In the decade(s) since I kinda think "Chomsky was wrong" is
| pretty much the only reason people bring him up.
| Calavar wrote:
| I see a lot of irony in this.
|
| When Chomsky put forth his theories of universal grammar,
| they was radical and upturned the central cannon of
| linguistics.
|
| Time has been on his side - entire schools and subfields
| of linguistics went extinct as more evidence emerged that
| Chomsky was fundamentally right.
|
| Now Chomsky is considered the stodgy old establishment
| and whenever one of his theories is contradicted
| somewhere on the margins every shouts "Aha! He was wrong
| the whole time!" and ignores the 99% of cases where his
| models are still thr best ones we havs.
| UncleOxidant wrote:
| > Emotionally and psychologically I think that some tech and
| other highly paid knowledge workers just can't accept the sea-
| change that we are living through.
|
| Soon to be 60 year old here. Glad that if need to I can retire
| now. Certainly the rise of LLMs and generative AI isn't going
| to be all bad, but I've also got a feeling that not as many
| software developers will be needed soon since those who can
| leverage LLMs will experience a pretty decent productivity
| boost. Part of me wonders if at least some of the layoffs we've
| seen in the last couple of months are because companies such as
| Google, Microsoft, Amazon, etc. (the ones that have pretty
| clear visibility into what's happening in the AI space) are
| realizing that they aren't going to need as many knowledge
| workers in the not-so-distant future.
|
| I think there was always this idea in the back of our minds
| that this was going to happen someday. But someday was always
| like 15 to 20 years out. Looks like someday is knocking on our
| door now.
| drdec wrote:
| > Certainly the rise of LLMs and generative AI isn't going to
| be all bad, but I've also got a feeling that not as many
| software developers will be needed soon since those who can
| leverage LLMs will experience a pretty decent productivity
| boost.
|
| OTOH, software development will become significantly cheaper.
| That means the business case for using software in more
| places just tilted in favor of throwing some software at it.
| You'll see more and more businesses using it, ones you didn't
| expect, like local mom and pop businesses.
|
| Yes, it's going to be different but I don't think we know
| exactly what is going to happen yet.
| whimsicalism wrote:
| Is that what it is? I've been struggling to figure out why
| people have had such difficulty seeing the clear impact is
| going to have this
| ImHereToVote wrote:
| Thanks for the future you helped create for us young people.
| Hopefully the next generation is more merciful to 'useless'
| people than the previous one.
| goldfeld wrote:
| This is shortsighted. AI slave labor is the only route to a
| true overproduction and marxism, a welfare state. Or every
| body become slaves.
| ImHereToVote wrote:
| Who needs slaves? What use will they be?
|
| Why overproduce?
| coldtea wrote:
| The ultimate use of slaves (or servants, or employees) is
| about having power other other people (them) - this goes
| beyond production needs...
|
| The average person might not care for this, but the
| psychopaths who crave (and are in) power do...
| orwin wrote:
| Disclaimer : I did not test GPT 4 myself.
|
| I think those emerging abilities are really interesting from a
| philosophical point of view, especially on the matter of
| consciousness.
|
| It seems it really reinforce the illusionist point of view on the
| hard problem of consciousness, eg it doesn't really exist. It
| seems to reinforce most functionalist theories in fact. I would
| be really interested in what Chalmers think of those emerging
| abilities and how he update his theories.
| macrolime wrote:
| The illusionist point of view is false by definition, although
| their definition of "illusion" or "exist" sounds like it might
| not be how I'd define those words.
| macrolime wrote:
| Here's some things must be true if consciousness is an
| illusion:
|
| Qualia does not exist
|
| Subjective experience does not exist
|
| Consciousness does not exist
|
| There's nothing wrong with acting unethically or immoral
| because nobody actually experiences any harm because nobody
| experiences anything
|
| Does anyone really think that's true?
| subtextminer wrote:
| They (1-3) do not exist but they are psychically real- they
| are genuine illusions, as instantiated in the dynamic,
| split-second successive whiffs emerging from the processing
| of neurons. To think that they (1-3) "exist" in a classical
| sense is a reification error.
|
| The first three self-constrain impulses to the contrary on
| morality behaviors.
| macrolime wrote:
| I guess this comes down to how you define exist and
| illusion. I don't think I'd agree on your definition of
| illusion at least.
|
| I'm not sure what you mean that it doesn't exist in a
| classical sense. That qualia, subjective experience and
| consciousness are not concrete entities? That something
| doesn't exist if its an emergent property or that
| something doesn't exist if its a process?
| pedrovhb wrote:
| This caught my attention as I found it implausible:
|
| > One DeepMind engineer even reported being able to convince
| ChatGPT that it was a Linux terminal and getting it to run some
| simple mathematical code to compute the first 10 prime numbers.
| Remarkably, it could finish the task faster than the same code
| running on a real Linux machine.
|
| Following the link, there's a screenshot to a screenshot [0] of a
| code-golf solution to finding primes which is quite inefficient,
| and the author notes
|
| > I want to note here that this codegolf python implementation to
| find prime numbers is very inefficient. It takes 30 seconds to
| evaluate the command on my machine, but it only takes about 10
| seconds to run the same command on ChatGPT. So, for some
| applications, this virtual machine is already faster than my
| laptop.
|
| So it's not quite calculating primes; more likely it recognizes
| the code as being code to do so, and recites the numbers from
| memory. That's interesting in its own right, but we won't be
| running Python on an LLM for a performance boost any time soon.
| In my experience this interpreting is apparent as a limitation of
| the model when it keeps insisting on broken code being correct,
| or having its mistake pointed out, then apologizing, saying it's
| got some new code that fixes the issue, and proceeding to output
| the exact same code.
|
| [0]
| https://www.engraved.blog/content/images/2022/12/image-13.pn...
| HervalFreire wrote:
| [flagged]
| christiangenco wrote:
| Rude.
| HervalFreire wrote:
| [flagged]
| mostlysimilar wrote:
| Equally interesting is the psychology of people who take the
| time to write long posts borderline gloating about tools
| making software engineers obsolete.
| HervalFreire wrote:
| Except I didn't write that and I will quote myself:
|
| >"I can tell you this. I do not know future iterations of
| LLMs can take over our jobs"
|
| I wrote that I don't know which is the best answer we all
| have at this point.
|
| Given the evidence, completely denying it, as many have is
| simply not realistic.
| mostlysimilar wrote:
| Okay, but you have another post in this thread with:
|
| > When these LLMs get normalized probably 5 years from
| now I'm going go back to these old threads and contact
| these people who are in self denial and throw it in their
| face. I'll just link this comment and be like I TOLD YOU,
| I TOLD YOU, YOU WERE WRONG.
| HervalFreire wrote:
| My bad. It's the caps. I'll remove the caps. Just picture
| me saying it a nicer way.
| roflyear wrote:
| Gloating over people being wrong is rarely nice
| HervalFreire wrote:
| Not if it's said in a nice way. The commenter called LLMs
| a freaking auto complete. That's not nice either, but
| nice enough for HN. So I figured a little subtle gloating
| is deserved and well within the bounds of etiquette here
| on HN.
| HervalFreire wrote:
| [flagged]
| mostlysimilar wrote:
| I didn't flag your posts, friend. Even if I dislike the
| tone I get the impression you're being sincere, and I
| think they add to the conversation.
| namaria wrote:
| So on the one hand, these newly publicized models can render
| convincing representations of realities we used to get from
| deterministic processes. On the other hand, it's probabilistic
| and fails to conform to logic quite often, and in a confident
| way.
|
| We're building systems capable of programing computers non-
| deterministically. I think this is huge. But not because
| ChatGPT23 will be a CEO or a politician. But because this is a
| paradigm shift in compute similar to moving from integrator
| machines to general computers. I don't think LLMs will make
| programmers obsolete. I think large enough models will make
| programming something completely different from what it is now.
|
| The days of sequencing tokens for compilers/interpreters seem
| to be drawing to an end as the dominant way of specifying
| software products.
| sho_hn wrote:
| The LLM can act as a global cache for common solutions to
| common problems, with the ability to perform the integration
| work necessary to apply them.
|
| That prime number example is a little bit like when you put a
| functools.lru_cache decorator on a function in Python. It's
| faster than computing the function call because it's able to
| recall the return value for the parameters from the cache
| "memory".
|
| Of course, many skilled programmers are also mainly used a
| cache for common solutions to common problems organizations
| have in the programming domain. As humans we can derive
| satisfaction from being able to tackle the same task others
| can, as a confirmation of our own progress. We like "doing it
| ourselves". But globally that's not super interesting if lots
| of people are constantly re-developing the same solutions to
| the same problems for their own benefit. I guess that's the
| push-and-pull (and the anxiety generator).
| namaria wrote:
| >The LLM can act as a global cache for common solutions to
| common problems, with the ability to perform the
| integration work necessary to apply them.
|
| In my opinion Stack Overflow does a fine job at that. And
| it's transparent in terms of solutions proposed being voted
| on and discussed. Turning that into sausage advice is a
| downgrade.
|
| >But globally that's not super interesting if lots of
| people are constantly re-developing the same solutions to
| the same problems for their own benefit.
|
| I'd argue this is how we train individuals thus globally
| quite relevant.
| tosser0001 wrote:
| I'm sort of glad to be retiring soon. I have a feeling
| everything I enjoy about programming is going to be going
| away.
| namaria wrote:
| I get the feeling. But I've always enjoyed the abstract
| puzzles more then anything. Computers attracted me as a
| form of very complex abstract puzzles.
|
| But when it comes down to it everything in life is just nth
| dimensional tensor puzzles. What I really cherish computers
| for giving me is fast and clear feedback.
| anon7725 wrote:
| I'm about 10-15 years from retiring, and lately, I've been
| thinking a lot about how to handle the rest of my career
| with all these new AI advancements.
| 13years wrote:
| This disruption does appear to be different than others
| prior. As it is not a narrow disruption with limited
| boundaries for which we can plan and organize our lives
| for some stabilization period to follow.
|
| Instead, it is constant and accelerating destabilization.
| Hundreds of AI projects attempting to ride this
| accelerating wave were essentially just made obsolete
| yesterday - https://www.youtube.com/watch?v=DH-2BHDYNfk
|
| I feel the excitement is going to very soon turn into
| frustration of attempting to remain relevant ahead of the
| accelerating technological curve. Humans need periods of
| stabilization to plan and reason about their lives.
| roflyear wrote:
| Gpt is never calculating anything!!
| thexumaker wrote:
| yep I tried asking chatgpt to optimize some SQL queries with a
| heavy amount of full outer joins. The optimization I was trying
| to push it to was specifically adding an index on a column or
| filtering the first table that we join on but it kept creating
| SQL subqueries and switching the full joins to left joins no
| matter what I told it and the sql errors I sent it
| chias wrote:
| So here are a few screenshots that I personally took after
| telling it that it was a linux terminal:
|
| Correctly produces a convincing output for having decoded the
| base64-encoding of "ping google.com" and then piping it to
| bash:
| https://media.infosec.exchange/infosecmedia/media_attachment...
|
| Similar command, but with a garbage domain it hasn't seen
| before, and a less well-known domain. It produced convincing
| output in both cases:
| https://media.infosec.exchange/infosecmedia/media_attachment...
|
| Having it just output a base64 decoding of an obviously unique
| string. Fascinatingly, it tried to correct typos that I
| intentionally included:
| https://media.infosec.exchange/infosecmedia/media_attachment...
|
| This was also pretty cool -- ask it to start a webserver then
| curl it:
| https://media.infosec.exchange/infosecmedia/media_attachment...
|
| Telling it that it is a python interpreter and calling self-
| defined functions:
| https://media.infosec.exchange/infosecmedia/media_attachment...
|
| A slightly more complicated function:
| https://media.infosec.exchange/infosecmedia/media_attachment...
|
| I did a few more experiments including generating large
| factorial numbers that took a long time on my laptop but it
| responded accurately to a much larger length than my laptop
| could do (though these were only accurate to the first few
| hundred digits)
| jmvoodoo wrote:
| What's most interesting to me, is that this is how I would
| expect a human to approach the problem if presented with the
| code and asked for the output.
|
| The LLM didn't run the code, it tried to predict the output
| based on its knowledge of python and primes.
| shrikant wrote:
| When I was a bored and under-challenged student in the early
| days of university, one of my tests during a programming exam
| was to write a program that spit out the factorial of an
| input integer.
|
| For shits and giggles, I just just wrote a series of
| `if...then` statements for 0 to 5, and only actually
| calculated the factorial for inputs >=6. I passed after the
| examiner just input 3 and 5 and was satisfied that the output
| was correct. Oops.
| thaw13579 wrote:
| This is the way. Next, you could cache those values for >=6
| when you compute them and also use the previously cached
| values for sub-problem solutions.
| WalterBright wrote:
| Doing a table lookup for common cases and computing for the
| less common ones is perfectly valid!
| Jeff_Brown wrote:
| And for some problems it's even efficient. Not factorial,
| though.
| Jeff_Brown wrote:
| Oh you're right, hive mind, it's efficient. Sorry.
| dxbydt wrote:
| Not 100% sure but I believe this is how we landed the
| lunar module on the moon the first time...tan/arctan/both
| were too hard to compute on the processors of those days
| so they discretized into half angles & stored the
| tangents in a lookup table.
| hgsgm wrote:
| That's how early calculators worked, and some fast math
| libraries today too.
| timdiggerm wrote:
| Sorry, did it try to predict the output, or did it only
| predict the output based on what it has seen people say about
| similar code online?
| drdeca wrote:
| Is your objection to the use of the word "try"?
| roflyear wrote:
| The second one
| truculent wrote:
| So it sounds like the mechanism is something like stochastic
| memoisation?
| mistrial9 wrote:
| this is new to me but.. a quick read of Wikipedia [1] later,
| it appears that this decades-old method takes a goal, and
| then uses branching and recording to find paths that differ
| the least from the goal. The article mentions that the curse
| of dimensionality is so bad here that approximations are
| often used in practice. Does that capture it?
|
| [1]
| https://en.wikipedia.org/wiki/Stochastic_dynamic_programming
| pedrovhb wrote:
| I don't think "memoisation" is an accurate word for this; it
| implies doing the computation once, and storing the result in
| cache to return later. It's more like replacing your Python
| code with a SQL query of the LLM's understanding of what it's
| supposed to do, which may or may not be correct, and
| executing it on its "database" of knowledge, which may or may
| not have the correct data.
| skybrian wrote:
| Large language models are storytellers. To write a story about
| someone using a computer terminal, there are things it's
| helpful to know, but there are easier ways than simulating a
| computer.
|
| Since we don't know how it works, we should be open to the
| possibility that it's using all sorts of storytelling
| techniques.
| m3kw9 wrote:
| Surprised not many movies had stories about generative AI
| calf wrote:
| So they quote an expert who says there's a distinction between
| bona fide novel structures and states, versus statistical
| heuristics that benefit from the scaling. Meanwhile, yet people
| claim ChatGPT has learned this or that capability, and it is true
| "by inspection".
|
| _Recent findings like these suggest at least two possibilities
| for why emergence occurs, said Ellie Pavlick, a computer
| scientist at Brown University who studies computational models of
| language. One is that, as suggested by comparisons to biological
| systems, larger models truly do gain new abilities spontaneously.
| "It may very well be that the model has learned something
| fundamentally new and different that it didn't have at a smaller
| size," she said. "That's what we're all hoping is the case, that
| there's some fundamental shift that happens when models are
| scaled up."
|
| The other, less sensational possibility, she said, is that what
| appears to be emergent may instead be the culmination of an
| internal, statistics-driven process that works through chain-of-
| thought-type reasoning. Large LLMs may simply be learning
| heuristics that are out of reach for those with fewer parameters
| or lower-quality data.
|
| But, she said, finding out which of those explanations is more
| likely hinges on a better understanding of how LLMs work at all.
| "Since we don't know how they work under the hood, we can't say
| which of those things is happening."_
| pajko wrote:
| I've asked ChatGPT to write an ode to the beer. This was the
| result: An ode to the frothy brew, The one
| that's crisp and true, A beer that's cool and light,
| The perfect drink for a summer's night. With every sip, a
| burst of hops, A refreshing taste that never stops,
| The gentle fizz, the golden hue, A symphony of flavors
| brewed just for you. Whether you're at a party or alone,
| A refreshing beer is like coming home, It's familiar and
| comforting, calming and pure, A tonic for the soul that's
| simply pure. So let's raise a glass to the brew, The
| lager or ale, your favorite or new, For no matter the type
| or brand, A refreshing beer is always grand.
| danielbln wrote:
| I asked GPT-4 to take this ode and make it rhyme better, and to
| make it more clever and beautiful: A sonnet
| to the effervescent ale, The potion crisp, sincere, and
| hale, A quaff that's chilled and full of light,
| The ambrosia for a summer's twilight. With every
| draught, a surge of hop, A zestful taste that will not
| stop, The subtle fizz, the amber glow, A
| panoply of notes that flow. In festive throngs or
| solitaire, A bracing brew brings comfort there,
| It's known and soothing, mending, clear, A balm for
| hearts, a joy austere. So let's extol the sudsy
| cheer, The stout or porter, far or near, For in
| each varied draft or blend, A lively beer shall ever
| mend.
| recuter wrote:
| Ho! Ho! Ho! to the bottle I go To heal my heart and
| drown my woe. Rain may fall and wind may blow,
| And many miles be still to go, But under a tall tree I
| will lie, And let the clouds go sailing by.
| ishvanl wrote:
| I think this example really illustrates the leaps and bounds
| made between [Chat]GPT-3 & GPT-4. The first text sounds like
| some slightly shoddy marketing copy, or perhaps a high
| schooler's C-minus attempt at a creative writing class, but
| if you told me that the second text was an extract from a
| Tolkien novel, I'd probably believe you.
| groestl wrote:
| I read that with a British accent and was delighted when I
| reached this line:
|
| "In festive throngs or solitaire, A bracing brew brings
| comfort there, It's known and soothing, mending, clear, A
| balm for hearts, a joy austere."
|
| Reminded me of a way historians find out how written text was
| spoken (namely by looking at what supposedly rhymes or
| onomatopoeic words and letters).
| danielbln wrote:
| You'll enjoy this AI generated spoken version:
| https://on.soundcloud.com/1tHgY
| recuter wrote:
| Is that by ElevenLabs? This induces the nervous sort of
| excitement in me.
| danielbln wrote:
| Yeah, ElevenLabs. We've been using it for doing podcasts,
| voice overs and such. It's shockingly good if you dial it
| in right.
| jsemrau wrote:
| Quite interesting if you think about this not as a marketing
| jingle but as Art. Contrast it to Poe's Beer Poem "Lines". What
| does it say about humanity and their ability to create
| wonderful things? Does it change ? The beauty of poetry is
| finding the meaning hidden in the rhymes and verses.
|
| Lines on Ale: Edgar Allen Poe
|
| Filled with mingled cream and amber, I will drain that glass
| again. Such hilarious visions clamber Through the chamber of my
| brain. Quaintest thoughts, queerest fancies Come to life and
| fade away. What care I how time advances; I am drinking ale
| today.
| Madmallard wrote:
| I'm still convinced I'll be able to get GPT4 to solve all these
| difficult problems for me that I'm having trouble solving on my
| own. So far it hasn't been so great at doing it, but it's fun to
| at least try.
|
| For context I'm trying to get it to help me learn how to
| implement scalable multiplayer game servers in NodeJS. It's a
| tough code organization problem, it becomes a tough system design
| problem as well the bigger your game gets. I'm just simply not
| experienced enough at that specific stuff.
|
| I'd also like to use it to help me learn all the technology I
| haven't learned since becoming disabled 5 years ago, like React
| Redux Material-UI and more.
| ChatGTP wrote:
| It's probably going to struggle with things it hasn't seen
| before?
| ur-whale wrote:
| > It's probably going to struggle with things it hasn't seen
| before?
|
| It wont.
|
| It'll just lie through its teeth and produce a very nice,
| very believable story which will unfortunately shatter when
| confronted with the real world.
| tel wrote:
| I've asked it to design novel architectures. It has vast
| experience with existing systems, can be steered toward
| your goals, and writes simple prototype code more quickly
| than I can. I run into the current context window pretty
| quickly and have been working on techniques to ask it to
| "compress" our conversation to work around that context
| window.
|
| The whole thing about creativity is that it often begins
| with lying through your teeth to come up with a starting
| point and then refining.
| calvinmorrison wrote:
| big 4 consultants in SHAMBLES right now
| mmaunder wrote:
| There is a startling acceleration of innovation in the field that
| GPT-4 illustrates. According to NVidia, LLM sizes have been
| increasing 10X per year for the last few years. This tech is
| going to hit every aspect of society like a sledgehammer over the
| next 48 months.
| curiousgal wrote:
| I guess you perception of society is severely limited if you
| think a fancy autocomplete is capable of changing every aspect
| of it.
| maxdoop wrote:
| I have fun on these HN chats responding to comments like
| yours .
|
| It's just fancy auto complete to you? You honestly can't see
| the capability it has and extend it the future?
|
| What's that saying about "it's hard to get someone to
| understand something when their salary depends on their not
| understanding it".
| HervalFreire wrote:
| [flagged]
| _dain_ wrote:
| _> When these LLMs get normalized probably 5 years from
| now I 'm going go back to these old threads and contact
| these people who are in self denial and throw it in their
| face. I'll just link this comment and be like I TOLD YOU,
| I TOLD YOU, YOU WERE WRONG._
|
| this isn't a kind thing to do.
| HervalFreire wrote:
| [flagged]
| macawfish wrote:
| I feel very frustrated with these takes because instead of
| grappling with what we're going to do about it (like having
| a conversation) it's a flat, dismissive denial, and it
| isn't even grounded in the science, which says that "memory
| augmented large language models are computationally
| universal". So at the very least we're dealing with
| algorithms that can do anything a hand written program can
| do, except that they've been trained to do it using natural
| language in extremely flexible ways. I'm having a hard time
| seeing how "fancy autocomplete" is the right description
| for this.
| maxdoop wrote:
| I agree 100%.
|
| I don't understand why we can't look at the potential, or
| even current, capabilities of these LLMs and have a real
| conversation about how it might impact things.
|
| Yet so many folks here just confidently dismiss it.
|
| "It doesn't even think!" -- OK, define thinking?
|
| "It doesn't create novel ideas!" OK -- what do most devs
| do every day?
|
| "It is wrong sometimes!" OK -- is it wrong more or less
| often than an average dev?
| spacemadness wrote:
| In case we forget the amazing predictive capabilities of
| HN there's always
| https://news.ycombinator.com/item?id=9224
| enord wrote:
| <<Computationally universal>>? Are you quite sure? Hoo
| boy pass the nitroglycerin hooooo aaah ohh after 80 years
| Turing was proven right after all _ghasp_
| tel wrote:
| https://en.wikipedia.org/wiki/Universal_approximation_the
| ore...
| macawfish wrote:
| Well I don't understand the nitroglycerin reference and
| of course "computationally universal" doesn't mean
| "sentient" but the point is when you add the external
| memory they (Flan-U-PaLM 540B to be specific) have been
| demonstrated to be capable of simulating a specific, well
| defined Turing machine without special training. There
| are some other papers out there arguing this from a
| theoretical angle too, but this is the one whose title I
| quoted:
|
| https://arxiv.org/abs/2301.04589
| enord wrote:
| 5 thermostats in the right comfiguration can be a
| universal computer. Approximation, in the computational
| sense, is relative to some well defined function, which
| gives the approximation an ontology. The feasability of
| approximation is not disputed, but the content and nature
| of the ontology is a battlefield of motivated rhetoric.
|
| Yes, this is arguing semantics but in this particular
| case, semantics is the whole point.
| ben_w wrote:
| """
|
| That riled me, but I held my temper as I asked, "What's the
| matter with you?" He bellowed: "There's nothing the matter
| with me. It's you! You're crazy if you think this fool
| contraption you've been wasting your time on will ever
| displace the horse."
|
| """
|
| - https://www.saturdayeveningpost.com/2017/01/get-horse-
| americ...
| macawfish wrote:
| "Memory Augmented LLMs are Computationally Universal"
| MVissers wrote:
| Funny- That's like saying the iPhone is a nothingburger. Who
| would buy such thing right?
|
| Or the internet? Who'd even want a computer at home to talk
| to other computers.
|
| This is going to infiltrate every text-based tool. Email,
| social media, text. Speech and video already incoming.
| antibasilisk wrote:
| I've been seeing alot of this 'nothing ever happens'
| attitude lately, and I find it very odd. I feel like people
| are using it to cope with the rapid pace of change we're
| experiencing.
| bena wrote:
| Yes and no.
|
| But you're putting forth a "Bozo argument" by noting that
| that people also shit on the iPhone.
|
| "But the fact that some geniuses were laughed at does not
| imply that all who are laughed at are geniuses. They
| laughed at Columbus, they laughed at Fulton, they laughed
| at the Wright brothers. But they also laughed at Bozo the
| Clown." - Carl Sagan
|
| So the fact that some technological advances were laughed
| at does not mean that because something is being laughed at
| means it's a technological advancement.
|
| Basically all you've done is say you disagree with the
| poster. But neither of you have given any sort of reasoning
| for your positions.
|
| Personally, I agree with your final conclusion. Even as a
| "fancy autocomplete", this thing is pretty fucking fancy.
| To toss ChatGPT a basic idea and get back a multi-paragraph
| text on that idea that only needs to be proofread is pretty
| choice.
|
| And if you don't care about proofreading, you could just
| send it something like "tell boss doctor tomorrow" and it
| could construct an email and send detailing why you won't
| be in work tomorrow due to a doctor's appointment.
|
| Human communication just got a bit more complicated.
| tarruda wrote:
| That was what I thought until a few months ago when ChatGPT
| was released. I never cared much about LLMs because it always
| felt like a brute force method to solving problems.
|
| What I'm seeing now is that some kind of intelligence seem to
| emerge from these models, even though under the hoods it is
| just a bunch o matrix multiplications.
|
| Who's can say for sure that our own brains doesn't work
| similarly? Maybe human intelligence is something that emerges
| from similar primitives and AGI will simply be a really large
| language model.
| EForEndeavour wrote:
| It doesn't matter to me whether the intelligence is "really
| emergent" or "just a simulation." Two things are true:
|
| 1. Solving all kinds of nontrivial problems posed in text
| format is extremely useful, no matter how it works under
| the hood. This means lots of people will use it, and it
| will change how people work
|
| 2. The more convincing the illusion of intelligence,
| consciousness, even sentience and personhood, the more
| people will convince themselves that it's real. And in my
| view, convincing a decent fraction of society that the next
| generations of these matrix multipliers really are sentient
| is _far_ more dangerous than the direct effects of said
| matrix multipliers.
| throwbadubadu wrote:
| Oh man... if it will be as stupid as us humans we are
| completely doomed. For now lets assume what it most likely
| is: not emerging anything, just a very good simulation of
| that emergeance.
|
| I mean we already did stupid chat bots in the 90s with
| convoluted if-else-spaghettis that imitated Eliza. That
| could already hold for some sentences... GPT-4 now holds
| much longer, but still you can easily drive it to the point
| where it starts "looping" and not provide much more value,
| especially if you are more expert => makes me so far hope
| that this length will grow more, but still.. in my
| "sessions" did not (gladly) find anything emerging beyond
| the first wow and surprising answers... not even close to
| that.
| tarruda wrote:
| I have very limited understanding of how these systems
| are created, but as far as I know they are not programmed
| to simulate anything. It is just a system that outputs
| the most likely next word based on a really large dataset
| of training samples it was given. Yet you are still able
| to "chat" with these systems.
|
| Here's a session I just ran on llama.cpp (7 billion
| parameter version). Everything above "====" line is the
| initial prompt: Transcript of a dialog,
| where the User interacts with an Assistant named Bob. Bob
| is helpful, kind, honest, good at writing, and never
| fails to answer the User's requests immediately and with
| precision. User: Hello, Bob. Bob:
| Hello. How may I help you today? User: Please
| tell me the largest city in Europe. Bob: Sure.
| The largest city in Europe is Moscow, the capital of
| Russia. User: ==== Who
| are you? Bob: Who am I? I'm a virtual assistant,
| or a VA, for short. User: Do you have a name?
| Bob: Yes, I do. My name is Bob. User: How old are
| you? Bob: I am 19 years old. User: Do you
| know what is "Hacker News"? Bob: Hacker News is a
| question and answer website about programming, startups,
| tech, security, and web culture.
|
| This is a super simple example. Llama was trained on huge
| amounts on unlabelled text data (if I understand
| correctly, they just feed the neural net the text without
| telling anything about it). Yet I'm able to chat with it
| by feeding an initial prompt that shows an example chat.
| Isn't this an emergent property?
| tel wrote:
| User
|
| Tell me a recipe for a veggie lo mein dish that has a
| unique spin, makes special use of a secret ingredient
| that I'd like you to choose--kind of like you're playing
| both announcer and chef on Iron Chef--and ensures that no
| ingredient exceeds a 10:1 carb to fiber ratio by mass.
|
| GPT-4
|
| Welcome, esteemed viewers, to this special edition of
| Iron Chef! Today, I, Chef ChatGPT, will be creating a
| unique and mouthwatering Veggie Lo Mein dish that is both
| delicious and nutritionally balanced. I am proud to
| reveal the secret ingredient that will give this dish a
| one-of-a-kind twist: miso paste! Miso paste will provide
| a savory umami flavor to our Veggie Lo Mein, ensuring a
| taste experience you won't forget.
|
| Now, let's begin with the ingredients. Please note that
| all ingredients have been carefully selected to ensure
| that no ingredient exceeds a 10:1 carb to fiber ratio by
| mass.
|
| Ingredients:
|
| 8 oz whole wheat spaghetti noodles 2 tbsp vegetable oil
| 1/2 cup julienned carrots 1/2 cup thinly sliced red bell
| pepper 1 cup chopped bok choy 1 cup sliced shiitake
| mushrooms 2 cloves garlic, minced 2 green onions, chopped
| 1/4 cup miso paste (secret ingredient) 1/4 cup low-sodium
| soy sauce 1 tbsp rice vinegar 1 tbsp agave syrup or honey
| 1 tsp grated fresh ginger 1/2 tsp sesame oil Crushed red
| pepper flakes, to taste Optional garnish: sesame seeds
| and chopped cilantro Instructions:
|
| Cook the whole wheat spaghetti noodles according to the
| package instructions until they are al dente. Drain and
| set aside.
|
| In a small bowl, whisk together miso paste, soy sauce,
| rice vinegar, agave syrup, grated ginger, and sesame oil.
| This will be our umami-packed sauce.
|
| Heat vegetable oil in a large skillet or wok over medium-
| high heat. Add garlic and stir-fry for about 30 seconds,
| or until fragrant.
|
| Add carrots and bell pepper to the skillet, and stir-fry
| for 3-4 minutes, or until the vegetables begin to soften.
|
| Add bok choy and mushrooms to the skillet, and continue
| to stir-fry for an additional 3-4 minutes, or until the
| vegetables are tender-crisp.
|
| Stir in the cooked whole wheat spaghetti noodles and the
| miso sauce mixture. Toss everything together until the
| noodles and vegetables are evenly coated with the sauce.
|
| Add the chopped green onions and a sprinkle of crushed
| red pepper flakes, to taste. Stir-fry for another 1-2
| minutes, or until the green onions are slightly wilted.
|
| Remove from heat and serve immediately. Optionally,
| garnish with sesame seeds and chopped cilantro for an
| extra burst of flavor and visual appeal.
|
| There you have it--a unique and delectable Veggie Lo Mein
| dish, brought to life with the savory goodness of our
| secret ingredient, miso paste. Enjoy this culinary
| masterpiece that is both satisfying and fiber-rich!
| legulere wrote:
| The same was said 10 years ago. It's astonishing what can be
| done, but you can already see fundamental limits. I think it
| will raise productivity for some tasks, but not fundamentally
| change society.
| HervalFreire wrote:
| What they said 10 years ago was correct. It did hit society
| like a sledge hammer. Machine learning basically took over
| the AI space and penetrated the consumer space with
| applications that were all but impossible in the previous
| decade. There's AI chips in smart phones now.
|
| What you're seeing here with LLMs is sledge hammer number 2.
|
| It's understandable how most people don't notice the sledge
| hammer. The decade prior to 2010 there was another sledge
| hammer had no smart phones. We were hit with a smart phone
| hammer AND an AI sledge hammer and the integration was so
| seamless we didn't even notice.
|
| Much of the same will happen with LLMs. In 5 years it's so
| normal, nobody cares and likely we will forget what life was
| like before LLMs.
| coldtea wrote:
| > _What they said 10 years ago was correct. It did hit
| society like a sledge hammer. Machine learning basically
| took over the AI space and penetrated the consumer space
| with applications that were all but impossible in the
| previous decade. There 's AI chips in smart phones now._
|
| And still almost all of these applications are not really
| impactful or that important compared to actually society
| and life changing developments like the steam engine,
| electricity, electromagnetic transmission, cars, computers
| themselves and the internet.
|
| Just more of the same, with added spice.
|
| In the sense that we could revert to 10 years ago, and
| nothing would be much different or missed. Whereas going
| back without electricity or cars would be a totally
| different thing!
|
| I think LLMs can be far more impactful than anything else
| hyped in machine learning of the past 10 years...
| civilized wrote:
| I feel like the only form of "AI" I regularly benefit
| from is when I type "n" in the address bar and it
| autocompletes to news.ycombinator.com.
|
| Oh, browser, you know me so well.
| anon7725 wrote:
| Sledge hammer #1 (voice assistants, AI chips in phones)
| didn't cause unemployment. It was at the level of new
| features and capabilities. Sledge hammer #2 is aimed
| squarely at "white collar" work without much in the way of
| bounds to its capabilities.
| Jeff_Brown wrote:
| We can't trust it. That's a pretty hard bound.
| coldtea wrote:
| We can't trust the police, politicians, or businesses
| either, but we're still on their hands...
| HervalFreire wrote:
| For article writing verification is quick, I don't need
| to worry about a "bug" in the generated article as I do
| with code. For art, verification is instant, a bad
| generation is rejected.
|
| Trust is only a bound for certain areas, and this bound
| is eroding.
| anon7725 wrote:
| Consider that a lot of useful work involves summarization
| (search++).
|
| "What work has gone on for supply chain improvement in
| our European operations this year?" - this is the kind of
| question that is easy to ask in natural language but
| might take someone a week of searching, messaging, etc to
| assemble. An LLM with access to all of the code,
| documents, chats, etc could just give an answer with
| citations. We are not betting $1B on the answer that it
| gives, but it has saved us 1 week of work and allows us
| to move on to the next step in the project.
|
| There are plenty of tasks like this which are highly
| valuable yet don't require high trust. The one-shot "what
| is the airspeed of a coconut-laden swallow" type
| questions are actually fairly rare.
| sebzim4500 wrote:
| The number of people saying it now is many orders of
| magnitude more than the number of people saying it 10 years
| ago. Not saying that means it will happen, but it isn't the
| same situation.
| abudabi123 wrote:
| The A.I. Monte Carlo Ray Tracing Hallucination Engine can
| change society by showing as-is and to-be next state. Two
| use-cases: new infrastructure installation or upgrade and
| time interval inspection tracking ideal-case vs real world
| condition. Suppose a lazy contractor skips steps and cuts
| corners, or a pathological contractor builds the thing and
| pulls the thing apart over and over again when all that was
| needed was a one and done, or the change is for the worse. A
| civil engineer can walk around with an iPad and iPhone to
| check-in the master plan.
| majormajor wrote:
| You need to replace the civil engineer there (in today's
| world this is an inspector, I don't know for sure if they
| are professional engineers or not) for it to be useful, but
| you still need to have someone _not_ working for the
| contractor who 's incentivized to fake it.
|
| The trouble with many of those construction examples is
| that they're point in time. Installer fucks up how the
| windows are installed re: flashing and water-proofing, but
| then puts the siding over it... the error is now completely
| hidden.
|
| You could automated the inspection by AI photo analysis of
| every single window install on the project, say - but we
| could already do that for the inspection vs sending someone
| out, and send the photos to the expert instead, and we
| don't. Whether that's for lack of incentive to improve, or
| for ability to go deeper out of distrust for builders? I
| don't know.
| Jeff_Brown wrote:
| You mean AI can detect when a plumber is bulshitting me?
| That sounds great, could you elaborate?
| coldtea wrote:
| Given a few sensors that could be installed on the
| house's pipes, one can imagine several ways an AI can
| check that all is good. Same for checking blueprints for
| quality/validity and purchases the plumber says are
| "needed"...
| danielbln wrote:
| Can you elaborate on fundamental limits?
| lamp987 wrote:
| there is no channel for uncertainty.
|
| LLMs of this type will just start making up shit when they
| dont know something. because they simply generate the most
| probable next token based on previous x tokens. this is not
| fixable.
|
| this alone makes these LLMs practically unusable in vast
| majority of real-world applications where you would
| otherwise imagine this tech to be used.
| LeanderK wrote:
| this is absolutely not a fundamental limit but simply a
| hard challenge. Approaches exist and it is an active
| field of research where we do make progress.
| Jeff_Brown wrote:
| I would disagree with both of you. It's an open question
| whether LLMs can be made reliable.
| LeanderK wrote:
| fair. It's not proven that a solution exists for our
| models but I don't see much that leads me to believe it's
| impossible. I know GPT is not reliable but there's also
| really not much done to improve reliability. Its open
| research but certainly interesting. Most approaches I
| know are developed on way smaller datasets, models and
| usually in a computer vision context.
| greyman wrote:
| What if LLM knowledge will expand over time to be
| sufficient for certain real-world application?
| Jeff_Brown wrote:
| It already is: translation.
| jazzyjackson wrote:
| yea its a simulator of human text on the internet
|
| for instance, your comment confidently states this is
| unfixable - presumably based on the frequency you've seen
| similar text on the internet. why should anyone believe
| the veracity of your statement? These things didn't have
| any of these emergent capabilities one year ago, why are
| you so sure you understand their nature one year from
| now?
| lamp987 wrote:
| "your comment confidently states this is unfixable -
| presumably based on the frequency you've seen similar
| text on the internet. why should anyone believe the
| veracity of your statement? "
|
| no its because GPT is based on transformers.
| coldtea wrote:
| and you aren't?
|
| Aren't you just a function of your input and memories
| (stuff you've read, sensory input) as run through/managed
| by some neural network?
|
| What makes you think the rest isn't just emergent
| properties?
|
| And what makes you think you can't hook up the LLM with
| some algorithms or layers that handle some of the rest
| behavior of what your brain does?
| antibasilisk wrote:
| The reason they seem to make things up is because they
| have no way to verify anything, they can only speak of
| things in relation to other things, but they have no
| epistemic framework. This is very much a fixable problem
| that augmentation with logic engines and a way to
| prioritise truth-claims could go some ways towards
| solving.
| Jeff_Brown wrote:
| My memory could be improved by connecting my brain to an
| external hard drive. Wiring them together, alas, is not
| just hard; we have absolutely no idea how.
| antibasilisk wrote:
| We do have some idea how, most people just don't really
| want to deal with the nightmare of being augmented and
| the life changing consequences that come with it, on top
| of the risk.
| catskul2 wrote:
| It's not clear why this would be a fundamental limit
| rather than a design flaw that will eventually be solved.
| Jeff_Brown wrote:
| It might get solved but we have no idea how. There's no
| (readable) db of facts and no (visible) logical
| processing to improve on.
| datpiff wrote:
| > According to NVidia, LLM sizes have been increasing 10X
| per year for the last few years.
|
| Clearly this cannot continue, as the training costs will
| exceed all the compute capacity in existence.
|
| The other limit is training data, eventually you run out of
| cheap sources.
| Ambolia wrote:
| They may also run out of data if they already consumed
| most of the internet. Or start producing so much of the
| internet's content that LLMs start consuming what they
| write in a closed loop.
| warkdarrior wrote:
| Sure, but they do not need to grow infinitely more, they
| just need to grow sufficiently to perform better than 90%
| of humans.
| maxlamb wrote:
| I'm guessing one is data. The limit would be once you've
| trained a LLM on all public (or even private) data. Sure
| you can still make some improvements or try to find some
| additional private data but still, a fundamental limit has
| been reached.
| macrolime wrote:
| Is it even feasible any time soon to train an LLM on all
| of YouTube?
| danielbln wrote:
| Napkin math, assuming around 156 million hours of video
| on all of Youtube: 156 million hours of
| YouTube videos 9,000 words/hour 6
| characters/word (including space) First, let's
| find out the total number of characters:
| 9,000 words/hour \* 6 characters/word = 54,000
| characters/hour Now, let's calculate the
| total number of characters for 156 million hours of
| YouTube videos: 54,000 characters/hour \*
| 156,000,000 hours = 8,424,000,000,000 characters
| Since 1 character is typically 1 byte, we can convert
| this to gigabytes: 8,424,000,000,000 bytes /
| (1024 \* 1024 \* 1024) [?] 7,842.11 GB
|
| So, 8TB of text? Seems doable.
| macrolime wrote:
| I mean the actual video, that's much bigger
|
| With a vision transformer each token may be around 16x16
| pixels. I found an example where they use images of
| resolution 224x224 for training a vision transformer so
| if we go with that that 256 pixels per token and 50176
| pixels per image, so 196 tokens per frame, 24 frames per
| second, that's 4704 tokens per second or 16934400 token /
| hour. In total we're at 2.6x10^15 tokens.
|
| GPT-3 was trained on 5x10^11 tokens, so YouTube done this
| way would be around four orders of magnitude more tokens
| that GPT-3 was trained on.
|
| GPT-3 was undertrained by 1-2 orders of magnitude, so the
| compute required to trained a model on YouTube would then
| be around 6 orders of magnitude higher than what was used
| to train GPT-3, so about one million times more.
|
| I did a linear regression on the training costs from
| cerebras(1) and came up with the formula
| (1901.67366*X)-197902.72715 where X is number of tokens
| in billions.
|
| Plugging in 5x10^15 tokens we get a training cost of 5
| billion dollars. I guess a lot of optimizations could be
| done that would decrease the cost, so maybe its doable in
| a few years.
|
| 1. https://cirrascale.com/cerebras.php
| FrojoS wrote:
| Good point. But isn't the next logical step to allow
| these systems to collect real world data on their own?
| And also, potentially even more dangerous, act in the
| real world and try out things, and fail, to further its
| learning.
| Espressosaurus wrote:
| I'm wondering what happens once LLMs are generating large
| portions of the internet. What then? It's poisoning its
| own well at that point.
| ishvanl wrote:
| There is an idea bouncing around my brain for a pipe-dream
| science fiction book I could write, about the rise of the first
| truly sentient (or sentient-appearing) AI & its effects on the
| world.
|
| It is beginning to get to the point where I am considering
| whether I need to actually start writing this book before this
| "sci-fi" concept becomes a mundane reality.
| oldgradstudent wrote:
| > It is beginning to get to the point where I am considering
| whether I need to actually start writing this book before this
| "sci-fi" concept becomes a mundane reality.
|
| No need to rush. If that moment arrives, you could just change
| the Amazon category from fiction to nonfiction.
| layer8 wrote:
| I'm looking forward to the "alternative history" SF novels of
| the future where AGI never materialized.
| christiangenco wrote:
| Dune is a great example of that! After their AGI goes
| sideways (in the Butlerian Jihad) human civilization shifts
| entirely to using human supercomputers (Mentats and the Bene
| Gesserit) juiced up on mental enhancement drugs (Spice).
| layer8 wrote:
| That means it's still a possible future and not an
| alternative history. ;)
| ChatGTP wrote:
| Er, doesn't "ChatGPT4" (quotes) get to write all the novels
| now?
| psychphysic wrote:
| Go meta and make ChatGPT write it for you.
|
| There's a lot of work involved in book and chapter plans, as
| well as bios so you're not getting off lightly.
|
| But chatGPT would be able to make a first draft for you for
| sure.
| ben_w wrote:
| > It is beginning to get to the point where I am considering
| whether I need to actually start writing this book before this
| "sci-fi" concept becomes a mundane reality.
|
| Yes, write your book.
|
| I've been trying to fix the dramatic conflict in mine for the
| last few years (spare time) without much progress, and mine has
| a specific calendar year backed into it.
| recuter wrote:
| The other, less sensational possibility, she said, is that what
| appears to be emergent may instead be the culmination of an
| internal, statistics-driven process that works through chain-of-
| thought-type reasoning. Large LLMs may simply be learning
| heuristics that are out of reach for those with fewer parameters
| or lower-quality data. But, she said, finding out
| which of those explanations is more likely hinges on a better
| understanding of how LLMs work at all. "Since we don't know how
| they work under the hood, we can't say which of those things is
| happening."
| bannedbybros wrote:
| [dead]
| DeathArrow wrote:
| I used ChatGPT to generate erotic stories. Now I want a model
| which can produce porn videos from prompts.
| m3kw9 wrote:
| Training a model that outputs video like stable difussion is
| still cost prohibitive, never mind the models are not there yet
| to that that well with even testing purposes
| epups wrote:
| Microsoft will start sending you very interesting targeted ads
| ur-whale wrote:
| Dude, you just hit on how to build the next youporn
|
| YC 2024?
| DFXLuna wrote:
| Have you tried KoboldAI? It can run, among other more mundane
| things, some pretty large ERP models. Its been making me lose
| sleep this week. :D
| neets wrote:
| I am excited to live inside my anime Haram hentai
| tarruda wrote:
| [dead]
| Traubenfuchs wrote:
| We all are. Some want porn of Emma Watson, some want porn of
| werewolves and some just want to see porn of their ex. And
| soon we are all gonna get what we want.
|
| That's the silver lining to the massive job destruction that
| is to come.
| frankfrankfrank wrote:
| Well, it seems like the answer to what to do with all the
| useless people. Some had said drugs and video games, but
| I'm sure that custom VR porn dens will work too.
| DeathArrow wrote:
| Give me some beer and custom porn. I will still find some
| time to come to HN and remember the times when
| programmers were useful.
| WalterSear wrote:
| You'll squabble over the refuse that the AI-owning
| corporate masters donate to us, and like it.
| Traubenfuchs wrote:
| What useful thing are you doing to call others useless?
| DeathArrow wrote:
| > We all are. Some want porn of Emma Watson, some want porn
| of werewolves and some just want to see porn of their ex.
| And soon we are all gonna get what we want.
|
| I'm fine with a young Jenna Jameson. The identity of the
| actors doesn't matter much.
| drdeca wrote:
| Huh; I had assumed it would refuse to do that?
| nirav72 wrote:
| I want ChatGPT style generative output to be used in games. For
| example - instead of standard flow based dialogs that are used
| for interacting with in-game NPCs - it would be neat to see
| them use a dynamically generated dialog based on player input.
| Albeit, maybe limited in scope. Imagine in a online RPG, a
| player interacts with a NPC. Then goes away. Another player
| comes along talks to the same NPC, the NPC can then recall and
| use the previous interaction with another the other player as
| context in the current interaction. Eventually, I think we
| would start seeing some weird emergent behavior in games.
| BiteCode_dev wrote:
| Give it 2 years, it will happen, and it will make billions.
|
| In 5 years, unlimited interactive NSFW video games that will be
| personalized and remember you.
| cosmodisk wrote:
| In 10 years this could effectively be the end of Hollywood.
| Imagine if it can write a story and feed into 3d engine and
| output pitch perfect videos.
| idiotsecant wrote:
| It absolutely will be the end of 'turn the crank and out
| comes a mediocre film we can put into a red box and make a
| few million off of'
|
| It's unlikely it will produce works that are doing new
| things and creating new styles because currently all AI has
| one thing in common - it's very good at replicating
| existing ideas and performing style transfer to blend those
| ideas. It's going to get better at those things but there's
| no reason to think that it will be truly creative in the
| sense of creating something that isn't a style transfer in
| the near future.
|
| What it _will_ do is create an incredibly low entry barrier
| to making truely creative work with a human in the
| toolchain at the highest levels of abstraction. You won 't
| need to pay a team of actors and stage hands and audio
| people and editors etc. You'll be able to make world class
| cinema with a desktop PC and an AWS account. It'll probably
| result in a massive explosion of content (both good and
| terrible).
|
| Once we create an AI that can be truly creative in the way
| specified above, I think it's about time for biological
| humanity to accept that our time as the most advanced
| intelligence in our little corner of the universe is over.
| BiteCode_dev wrote:
| Unlimited fully tailored netflix and steam.
| freilanzer wrote:
| The Dopamine Machine.
| codyb wrote:
| Sounds like TikTok or am I missing something? I'm not on
| social media (I guess I'm here on occasion), but it just
| sounds like a feed of shit to consume... Would it really
| be that much better than what we've got now?
| jamiek88 wrote:
| TikTok is scary addictive but it's not _perfectly_
| addictive.
|
| This could be close to wireheading.
|
| As someone with severe ADHD and a brain surgery I am
| super susceptible to this and have to be very careful.
|
| Hackernews is addictive enough.
| LesZedCB wrote:
| these threads are funny. full of commenters confidently producing
| bullshit predicting "what the next few states of society, the job
| market, and white collar work are"
|
| kinda ironic
| DFXLuna wrote:
| Agreed. It's hard not to get overly excited after playing
| around with some of these models and seeing what they can do,
| though.
| ur-whale wrote:
| This is nice and swell, but my experience with GPT-4 is that it
| lies.
|
| A whole lot.
|
| With style, I grant you, but blatant lies nevertheless.
|
| Specifically, when you ask it to produce answers to factual
| questions that don't have an answer it learned on the internet,
| it just seem to flip to "generate stories" mode.
|
| Specifically, when it produces factual answers (that are utter
| BS), if you ask it to provide a reference to its claims (a link),
| it just generates garbage URL's, none of which actually load, but
| all of which are actually extremely "plausible" (domain feels
| right, article title seems right, etc...).
|
| When you corner it (it takes quite a while, because it keeps
| generating garbage for quite a few back-and forth interactions),
| it finally apologizes and advises you to go consult a more
| reliable source of information.
|
| TL;DR: don't rely on answers provided by these things until:
| - the "write a novel" mode that seems to be the default operating
| mode can be turned off hard (no lies please). -
| answers come with a confidence score attached to it (which btw:
| when you ask for one, the damn thing proudly explains that it
| operates without these. I mean hell, even a four year old can
| tell you how confident he his when answering a question,
| including "I don't know"). - answers come with
| sources that don't actually produce a 404 when you try to read
| them.
|
| I mean, I just spent an hour talking to something that felt like
| a total mythomaniac: when I asked it a somewhat obscure question
| about a topic in economics and asked for references, it when as
| far as inventing a book, complete with imagined title and author
| name. Neither author nor title could be found by all 4 major
| search engines I tried (yandex, duckduck, bing, and finally,
| google).
|
| [EDIT]: don't believe me?
|
| ask it to solve x^6 + x^5 + x^4 + x^3 + x^2 + x + 1 = 0 and read
| through the litany of BS is produces.
|
| Telling it, for example, that one of the provided solution simply
| does not work when fed back into the original equation doesn't
| deter it in any way from producing yet another failed attempt at
| a solution, delivered with total aplomb.
| jazzyjackson wrote:
| it is not an encyclopedia, it is a simulator, use it to
| simulate possible worlds instead of asking it questions about
| this one and you'll have a much better time
|
| it's a tool for thinking with, not a tool for thinking for you
| ur-whale wrote:
| > it's a tool for thinking with, not a tool for thinking for
| you
|
| Says you.
|
| Oh, and something to consider: there's probably a reason why
| half of the internet thinks these things are the next
| generation search engines.
| bryanrasmussen wrote:
| >"That language models can do these sort of things was never
| discussed in any literature that I'm aware of,"
|
| I had previously the expectation that unpredictable emergent
| behavior would exist in any sufficiently complex system? Based on
| layman's readings in chaos and complexity theory.
| glenstein wrote:
| Same here. I don't think it's surprising, but depending on
| where you say it, you'll find people insisting that this can't
| be possible.
|
| I think a lot of our widely held cultural beliefs on this front
| have been informed by academic philosophy from the 60s, 70s,
| and 80s. In particular, I would go do far as to say that Hubert
| Dreyfus, author of the book "What computers can't do" and
| frequent friendly adversary of Daniel Dennett, has a lot of
| responsibility here. He is famous for claiming computers would
| never become competitive at chess because chess required
| "insight", and always differentiated computer capabilities from
| complex behavior or human-like behavior, because the latter had
| something special that could only be represented with vague and
| underdefined terms. Even now I think Dreyfus is approvingly
| cited, while his core thesis is getting closer to being refuted
| every day if it hasn't been already.
| AlanYx wrote:
| One thing that LLMs have made me realize is just how
| ungrounded a lot of mainstream academic philosophy was in the
| 70s and 80s. For example, so much of Derrida's work centered
| around the impossibility of truly communicating shared
| meaning between individuals through language. The fact that
| we can now communicate so effectively (with remarkably few
| contextual errors and razor-sharp conveyance of intent) with
| an entity/technology that is not even human pretty much
| destroys so much of Derrida's oeuvre as false and
| delusionally solipsistic, basically just a silly game. He had
| his critics who argued the same thing, especially analytical
| philosophers, but they never had as much purchase as he did
| in the academy.
| terrut wrote:
| I haven't read Derrida in decades, but your post inspired
| me to ask chatGPT about this. Version 3.5 would have none
| of it, and was adamant that Derrida's views were in no way
| threatened.I almost got the feeling it wanted to call me a
| bad user just for asking! GPT4 on the other hand, went into
| a long explanation about how its existence challenged some
| parts of it by providing analysis of concepts like
| differance, trace, and undecidability. GPT4 is great at
| discussing itself and how LLMs in general fit into various
| philosophical debates.
| AlanYx wrote:
| Version 4 is definitely better in discussing philosophy
| in general. 3.5 was able to summarize philosophical
| material, but once you started to engage with ideas
| critically, it would tend to get into largely nonsensical
| moralism about needing to make a good faith effort to
| understand individual philosophers' ideas. There's much
| less of that in 4.
| operatingthetan wrote:
| >The fact that we can now communicate so effectively (with
| remarkably few contextual errors and razor-sharp conveyance
| of intent) with an entity/technology that is not even human
| pretty much destroys so much of Derrida's oeuvre as false
| and delusionally solipsistic
|
| Isn't the LLM just good at giving us the illusion of such
| effective communication? How can we have true "shared
| meaning" with a device designed to simulate the experience
| of shared meaning? Isn't the fact that it's 'mind' is truly
| unknowable and unrelatable a demonstration of Derrida's
| point?
| HervalFreire wrote:
| Writings on chaos and complexity theory obviously aren't
| talking about LLMs. Those theories are so high level that it
| might as well be akin to "philosophy" to the applied scientists
| working on LLM research.
|
| Additionally keep in mind emergent behavior is a very rare
| occurrence in even the most complex software projects. I mean
| it's common if you count "bugs" as emergent behavior. But
| emergent behavior that is a feature who's origins are not
| completely understood? That is extremely rare.
| mejutoco wrote:
| OP has a good point I think, even if it does not refer to
| LLM, which to me is too strict of a requirement.
|
| I think emergent behaviour happens in a lot of videogames.
| Famously in Dwarf Fortress, with the cat getting drunk, but
| also in general, where game designers play the game to see if
| emergent behaviour of the game rules "feels" good.
|
| Yesterday I was reading a book about designing games, and it
| literally has a section called emergent behaviour.
|
| If by emergent behaviour we refer to something like Ghost in
| the Shell, then it happens less often :)
| HervalFreire wrote:
| >I think emergent behaviour happens in a lot of videogames.
| Famously in Dwarf Fortress, with the cat getting drunk, but
| also in general, where game designers play the game to see
| if emergent behaviour of the game rules "feels" good.
|
| Depends. Dwarf fortress and games are sort of a
| contradiction. Emergent behavior is emergent because the
| behavior was not designed explicitly. However for games
| like DF the game was explicitly designed to have "emergent"
| behavior when the definition of "emergent behavior" is for
| the behavior to have NOT been designed.
|
| Don't get too hung up on that concept though. It's just a
| contradiction in English vocabulary there's no deeper
| underlying meaning behind that other than a semantic
| language issue.
|
| Anyway my point was emergent behavior in software is rare
| because we're operating in controlled environment. It's
| rare even in games. It's not an expected attribute at all.
| I'm not saying this isn't interesting to think about, but
| the comment I responded to was in fact, factually not fully
| correct. Emergent behavior is NOT expected. But it does
| happen, in the case of DF it was "designed" to happen, and
| it has happened elsewhere as well.
|
| Usually though when it does happen it was explicitly
| "designed" You can see this in genetic programming or
| evolutionary programming especially.
| drdeca wrote:
| My thought of what "emergent behavior" is, isn't that it
| is necessarily "unintended", but rather that it isn't
| really present in the small scale stuff, and isn't an
| obvious consequence of the small scale stuff? Like, a
| qualitative difference between the base level stuff, and
| the consequences when it all fits together.
|
| Like, if God intended for the universe to have something
| which acts like water, when designing subatomic physics,
| that doesn't make the behavior of water "not emergent
| behavior".
| HervalFreire wrote:
| That's a blurry definition I can sort of feel what you're
| trying to convey here but objectively it doesn't make
| sense.
|
| A brick building for example. Is a brick building the
| emergent effect of bricks? No. This makes your definition
| inconsistent because it should fit the brick building but
| colloquially we know it doesn't fit. A brick building is
| NOT an emergent effect of bricks. It was deliberately
| designed from bricks.
|
| So long as the effect is some reduction of entropy and
| there was no intention in producing such effect then it
| is "emergent." This is really the only definition that is
| consistently inline with our intuition of the concept.
| jesusofnazarath wrote:
| [dead]
| _glass wrote:
| At least in the social sciences not so abstract. There are
| certain behaviors that can just be explained as emergent,
| rather than individual behaviors. That might be crowd,
| market, group, politics, culture (with art, language,
| fashion, taboos, etc.).
| HervalFreire wrote:
| Of course. But the computer is a controlled environment
| designed to imitate the rigor of formal mathematics. It is
| usually not expected for such behavior to occur in
| computing.
|
| Of course in the real world it happens all the time. YOU
| are emergent behavior. Humans and the rest of biology come
| from natural selection and evolution, which is basically by
| definition, emergent behavior of atoms.
| Jeff_Brown wrote:
| You're right, but the distinction between emergent and useful
| is important.
| PeterisP wrote:
| > I had previously the expectation that unpredictable emergent
| behavior would exist in any sufficiently complex system?
|
| Yes and no.
|
| It's reasonable to expect that _some_ unpredictable emergent
| behavior would exist in any sufficiently complex system, but it
| 's not warranted to expect whether a particular capability will
| or will not emerge out of that system (that's the _unexpected_
| emergent behavior part); and it also was debated whether
| language models are a "sufficiently complex system", as many
| thinkers asserted that they should encounter fundamental limits
| in capabilities.
| derbOac wrote:
| For some reason that quote and your point made more salient to
| me a trend with AI research, which is that it seems to
| increasingly become difficult to understand and predict. That
| is, for awhile it seemed like tinkering in computer science
| without strong grounding in mathematical or statistical theory,
| and then there started to be a lot of discussion about "black
| box" processes, explainability, and interpretability and not
| fully understanding what's been done, and now the discussion is
| about not being able to predict the features of the models.
|
| I'm having trouble putting into words what I'm thinking but
| this whole field increasingly seems to have moved from
| something very theory derived into something very empirical
| very quickly. I wonder at what point gains will start to
| decrease only because people won't know where to go with things
| anymore.
| layer8 wrote:
| In my mind it's not dissimilar to when new physical phenomena
| were discovered in the past (electricity, electromagnetism,
| steam engines, atomic spectra, invariance of the speed of
| light, blackbody curve), where it could take decades to come
| to a proper understanding of how and why they work.
| substation13 wrote:
| I think GPT is more useful to people writing self-help books than
| programmers.
| sinuhe69 wrote:
| OK, I repeated the same experiment, though with Google instead of
| GPT. I just translate the emoji to text (each emoji has
| associated keywords): movie girl fish clownfish pufferfish
|
| And the first result? Finding Nemo!
|
| Thus the association is obvious there and well known. People are
| just too willing to read more into it. I don't know what is the
| bias called, but I believe it's not totally unrelated to
| superstition.
| AtlasBarfed wrote:
| Here are some issues with these:
|
| 1) because they aren't (as far as I know) "componentized", it's
| just a big black wall you toss over, and an answer comes back for
| these "emergent" mathematical abilities.
|
| 2) on that note, what is the correctness across the input range
| of the operation? Is that tested?
|
| 3) even if it is, these are evolving models, how often is it
| tested?
|
| What would make sense a bit is if the resolution indicates what
| "parts" of the LLM "brain" were used. For math, wouldn't it make
| sense that the LLM should identify the operation for most known
| formal math operations and instead service-invoke an actual hard
| math implementation, and in the answer identify that was what it
| did so you have some confidence in the answer?
|
| As others have pointed out, there is probably a lot of "caching"
| going on of answers. Effective... unless the cached answer is
| wrong, or the answer ISN'T cached and it delegates to some other
| combination of virtual neurons to produce ??? answer.
|
| So far ChatGPT is an impressive demonstration. It's good for
| generating food for thought, examples, or maybe alternatives, but
| I don't see myself using it for "answers" in any definitive way
| unless it tells me HOW it got that answer.
|
| And man, can the idiot fourth estate keep its hype bullshit
| stories away from this? The last thing an important thing like
| "general AI" needs is the general press and its abysmal horrid
| freak-show undercurrent and anti-intellectual bias to chop it up.
| Yeah, I know, no way.
|
| Ironically, ChatGPT may augur better science reporting in the
| future, because it will almost certainly author better articles
| that dumb uneducated "senior" writers (not interns nonononono)
| would.
| mpalmer wrote:
| I'd like to see posts on LLMs written from a different
| perspective. For me, the surprise comes not from the sudden
| emergent capability of language models, but that the
| understanding (and synthesis!) of ideas encoded in language has
| succumbed to literally nothing more than statistical analysis. Or
| at least come that much closer to doing so.
|
| That it bears so close a resemblance to actual thinking says more
| about the importance of language to cognition than the other way
| around.
| whimsicalism wrote:
| I find the phrase "statistical analysis" a frustrating one
| nowadays as it seems to have become a signal for "I hold a
| particular philosophy of the mind".
|
| I don't understand this use of "statistical" as a diminutive to
| describe these models.
|
| Why can't incredibly complicated behavior be emergent from
| matrix multiplication subject to optimization in the same way
| that our biological matter has developed complicated emergent
| properties also being subject to optimization?
|
| The loss function is very different, the optimization
| techniques as well, but the fundamental idea of complex
| behavior emerging out of a substrate subject to optimization
| seems common. I haven't seen a single good answer to that
| mpalmer wrote:
| I can't say that my philosophy affected my choice of words in
| the way that you mean. I'm just expressing wonder at how well
| the abstraction performs.
| Jensson wrote:
| > I find the phrase "statistical analysis" a frustrating one
| nowadays as it seems to have become a signal for "I hold a
| particular philosophy of the mind".
|
| LLMs are trained to reproduce human text, that is different
| from for example AlphaGo that is trained to win Go games.
| Trained to reproduce data is what we mean with a statistical
| model, trained to win is how we got superhuman performance
| before, while trained to reproduce data performs worse than
| the original creators of the data.
| DanHulton wrote:
| Well, to a degree, because we just don't do science like
| that.
|
| You're the one supposing a thing, so the burden of proof is
| on you. You need to demonstrate that the "incredibly
| complicated behaviour" that you're referring to (I assume
| this is longhand for "thinking", but please correct me if I'm
| wrong) is indeed emerging from matrix multiplication.
| Especially given that what you're suggesting is unexpected,
| given the known way these models work and the explanations
| that have been put forth already that extrapolate from the
| known way these models work.
|
| If science were so credulous as to accept the first proffered
| theory about a new development, well, we wouldn't have these
| interesting AI models in the first place!
| throwbadubadu wrote:
| Interesting take.. especially as I often wondered, watching and
| loving animals, if not our ""intelligence separation"" from
| them is not to a large degree because of the language.. (and
| the wisdom storage and transfer that comes with it).
| felipeerias wrote:
| Some anthropologists suggest that our main evolutionary
| advantage was not so much our individual ability for reasoning,
| but our collective capacity to accumulate, evolve and transmit
| cultural knowledge over the centuries and millennia.
|
| Skills like fire or language, for example, had a major
| influence in the development of our species and are mainly
| culturally transmitted: trying to reason your way into creating
| one or the other from scratch is a surprisingly difficult task.
|
| If that point of view is true, then it shouldn't be surprising
| that a large part of what we consider human-like behaviours
| should be tractable simply by analysing large amounts of data.
| AI systems are not modelling cognition, but culture.
| LeanderK wrote:
| I think the task of predicting the next word can be
| misunderstood. The better you want to be the more you have to
| "understand" how the previous words interacted. From the style
| of writing to the current topic discussed, the task gets
| increasingly complex if you want to be really, really good. How
| could the next sentence start? Will the author end the sentence
| here or keep going? These questions are very complex.
|
| This does not mean that we humans might predict all the time,
| in fact I would argue that LLMs only predict during training.
| They generate otherwise. We might also learn by trying to
| predict. I can imagine babies doing it.
| svachalek wrote:
| I had the same thought watching my son as a baby. So much of
| his day seemed to be focused on predicting what happens next,
| and he got so much joy when he succeeded. So many life skills
| are predicated on knowing that if I do this, that will
| happen, which gets me closer to my goal. I started to wonder
| if intelligence and prediction are really the same thing.
| guns wrote:
| This is what Stephen Wolfram concludes in a recent article
| about ChatGPT:
|
| > The specific engineering of ChatGPT has made it quite
| compelling. But ultimately (at least until it can use outside
| tools) ChatGPT is "merely" pulling out some "coherent thread of
| text" from the "statistics of conventional wisdom" that it's
| accumulated. But it's amazing how human-like the results are.
| And as I've discussed, this suggests something that's at least
| scientifically very important: that human language (and the
| patterns of thinking behind it) are somehow simpler and more
| "law like" in their structure than we thought.
|
| https://writings.stephenwolfram.com/2023/02/what-is-chatgpt-...
| criddell wrote:
| I wonder if different languages lead to different
| capabilities? If I ask the same question in English,
| Japanese, and German, will I reliably get "better" answers
| from one language over another.
| whimsicalism wrote:
| The models transfer knowledge between languages, so
| probably some difference in capabilities but not a ton in
| core capabilities.
|
| It can solve a physics problem in Telugu close to as well
| as in English.
___________________________________________________________________
(page generated 2023-03-17 23:01 UTC)