[HN Gopher] Large Language Models As General Pattern Machines
       ___________________________________________________________________
        
       Large Language Models As General Pattern Machines
        
       Author : optimalsolver
       Score  : 69 points
       Date   : 2023-08-20 10:12 UTC (12 hours ago)
        
 (HTM) web link (arxiv.org)
 (TXT) w3m dump (arxiv.org)
        
       | asklotsogque wrote:
       | Amazing how much the mainstream media are still obsessed with
       | LLMs and factual accuracy. I prefer to see them as the worlds
       | greatest innovation as lateral thinking machines ie pattern A
       | applied over pattern B to give a credible pattern C.
        
         | hackernoteng wrote:
         | I think there is more innovation to happen with LLM output, for
         | example leveraging the attention weights to give information on
         | what the key events in history should be paid attention to
         | given a sequence of current events, etc. More predictive
         | information beyond just generated text responses
         | ("completions").
        
           | monkeydust wrote:
           | A useful feature of RAG is actual text citations, found it
           | useful in steering the user to develop their own reading list
           | for further follow ups
        
         | martindbp wrote:
         | Remember the outcry against using Wikipedia as a source in
         | school-work etc? Pretty convinced that the fear that "LLMs just
         | make stuff up" will gradually go away as they get better.
         | 
         | Also, at the moment you should view e.g. ChatGPT as your
         | autistic well-read friend who really do know quite a lot of
         | things, but only approximately, and is prone to make things up
         | instead of being found out for not knowing. It's OK to ask your
         | autistic friend how electricity works, but don't expect him to
         | correctly cite the titles and authors of relevant research
         | papers. Just common sense stuff if you really think of the LLM
         | as a brain of compressed information and not a search engine.
        
         | curtisf wrote:
         | LLMs are, and will continue to be, used for applications where
         | factual accuracy is important. The decisions that LLM outputs
         | prompt will soon be having big impacts in people's actual
         | lives, if they aren't already.
         | 
         | The appearance of credibility of the output makes this much
         | worse, since inevitably decisions will be deferred to LLMs that
         | they are not sufficiently accurate enough for.
        
           | hackernoteng wrote:
           | Real world industry wont stand for "hallucinated" outputs,
           | unless there is more innovation around UI/UX on outputs. For
           | example, no way lawyers/bankers/doctors are going to use LLM
           | in their current forms and limitations if they can't trust
           | the outputs.
        
             | JPLeRouzic wrote:
             | If you have employees, you also need to create transverse
             | structures ( _make people work in teams, set check lists,
             | QA, HR, accounting, create corporate charters on gender
             | equality and many other topics, tell many "statements that
             | ..." or "our mission is ...", create corporate culture, _)
             | because you can't trust humans employees at 100%.
             | 
             | Actually some of banks biggest failures were when only a
             | few and in some cases only one person was in charge.
             | 
             | https://en.wikipedia.org/wiki/List_of_corporate_collapses_a
             | n...
             | 
             | How is that different from LLMs?
        
             | captn3m0 wrote:
             | Professional norms differ between countries. What might be
             | unthinkable for a doctor in US might be meh for a
             | professional in India.
             | 
             | For eg, a judge in India used ChatGPT for a bail hearing.
             | https://www.livemint.com/news/india/this-indian-court-has-
             | us...
        
             | npsomaratna wrote:
             | Disagree. As a lawyer I use LLMs with RAG to help me
             | surface information all the time. Often, this allows me to
             | find niche case law that I just wouldn't have had the time
             | to find on my own. However, I double-check everything, and
             | read all the original sources.
             | 
             | LLMs are best treated as the AI equivalent of a human
             | assistant who is knowledgeable and fast, but also
             | inexperienced, and thus, prone to making mistakes. You
             | won't throw out the work of such an assistant-it'll still
             | save you hours of effort. However, you won't take the work
             | at face value either.
        
               | squeaky-clean wrote:
               | I have a friend in law school at the moment, and while he
               | obviously can't use AI for school, multiple professors of
               | his have recommended he get familiar with using them now
               | so he'll be efficient at using them after passing the
               | bar.
        
               | jebarker wrote:
               | I use LLMs for helping with coding it's proving
               | invaluable. I see it very similarly, an inexperienced
               | assistant with very broad knowledge. I also find that
               | they don't make too many mistakes in tasks such as
               | refactoring or finding bugs, it's when you ask it to just
               | wholesale generate code for you that you hit problems. If
               | I were to just take the code as is, not test it and not
               | check I understand it then use it it's me that would be
               | making the mistake not the LLM.
        
               | arrowsmith wrote:
               | What's RAG?
        
               | simonw wrote:
               | Retrieval Augmented Generation.
               | 
               | It's the trick where you take a question from a user,
               | search for documents that match that question, stuff as
               | many of the relevant chunks of content from those
               | documents as you can into the prompt (usually 4,000 or
               | 8,000 tokens, but Claude can go up to 100,000) and then
               | say to the LLM "Based on this context, answer this
               | question: QUESTION".
               | 
               | I wrote about one way to implement that here:
               | https://simonwillison.net/2023/Jan/13/semantic-search-
               | answer...
        
             | gryn wrote:
             | funny you mention lawyers when just not long ago some of
             | them made the news because they trusted chatGPT which gave
             | them a imaginary precedent.
             | 
             | just because a small percentage technically oriented people
             | know the limitations of LLMs doesn't mean that the rest do.
             | 
             | people have a tendency to anthropomorphize things, so they
             | naturally think that LLMs think in the same way other
             | humans they know think.
             | 
             | edit: first link I found about it, there was also some
             | posts about it here in HN.
             | https://www.abc.net.au/news/2023-06-09/lawyers-blame-
             | chatgpt...
        
             | naasking wrote:
             | > Real world industry wont stand for "hallucinated" outputs
             | 
             | Of course they will, if the other benefits are large
             | enough. Checking factual accuracy of a large corpus can
             | often be considerably simpler than generating the large
             | corpus to begin with.
        
             | ryanklee wrote:
             | Real world industry already relies heavily on hallucinated
             | outputs. They are called humans.
             | 
             | One of the main differences between people who see the
             | astonishing value of LLMs right now and people who are some
             | combo of skeptical, dismissive, and indignant, is the
             | expectation that for something to be a valuable source of
             | information, it has to be factually accurate every time.
             | 
             | The entirety of civilization has been built on the back of
             | inaccurate sources of information.
             | 
             | That will never change and it absolutely can't, because (1)
             | factual accuracy is not something that can be determined by
             | consensus in a variety of significant cases, and (2)
             | factual accuracy as a concept itself does not have a
             | consensus definition, operationally or in the abstract.
             | 
             | Absolutely frustrating to see these topics addressed as if
             | thousands of years of intense thinking around truth and
             | factual accuracy has not taken place.
             | 
             | The results of those inquiries do not support the basic
             | assumptions of these conversations (i.e. that factual
             | accuracy is amenable to exhaustive algorithmic
             | verification).
        
           | BoorishBears wrote:
           | That's a product problem, not an ML problem.
           | 
           | If I'm making a tool for doctors, I don't need to surface the
           | exact scientific fact the LLM recalled from memory, I can
           | design an interface that surfaces verbatim text from sources
           | based on the LLM's understanding of the situation.
           | 
           | No serious product should be surfacing a ChatGPT style chat
           | window to the user, it's a poor UX anyways with awful
           | discoverability.
        
           | flangola7 wrote:
           | But how long will factual accuracy remain below human level?
           | I expect the incorrect information issue to be a short term
           | problem. That new GPT-4 based legal AI the BigLaw firms are
           | signing up for already produces some documents with a lower
           | error rate than a human lawyer.
           | 
           | Perfection isn't necessary, it only needs to be better than
           | the average human. In a year or two I expect those kinks will
           | be worked out for countless job tasks.
        
       | golol wrote:
       | This is the real breakthrough of LLMs imo. Karpathy had a series
       | of tweest about this a year ago. It makes sense: Prettained LLMs
       | learn to continue sequences of tokens. These are language tokens
       | but the language is so complex and rich that a kind of general
       | pattern recognition and continuation ability is learned. I think
       | finetuning and alignment destroys a large part of this in-context
       | learning, so we are currently not focusing on this ability. In
       | the future we will train on all kinds of token sequences.
        
         | api wrote:
         | If we can analyze these machines deeply and understand how they
         | are doing this, is it possible that we can extract some kind of
         | general purpose pattern recognition and manipulation algorithm
         | or set of algorithms?
         | 
         | Are we just inductively generating algorithms that could in
         | fact run more powerfully and efficiently if they were
         | "liberated" from the sea of matrix math in which they are
         | embedded?
         | 
         | Or is the sea of matrix math fundamental to how these things
         | work?
        
           | golol wrote:
           | I don't think there is a magic alrogithm at play. I think
           | it's all about studying the data - the geometry of the data
           | manifold. In the case of a transformer more precisely
           | learning the geometry of the stochastic process of tokens.
           | Common "patterns", such as repetition or symmetry, appear
           | there and can be learned. I think like learning eigenvectors
           | in PCA you just learn more and more patterns and combinations
           | of patterns. Then at some point that starts looking like
           | pattern recognition.
        
           | jncfhnb wrote:
           | The sea of matrix math is the fundamental way these things
           | work. You could grossly simplify individual patterns to small
           | explicit algorithms but this is the way for general patterns.
           | Compress your information, and then synthesize its new
           | information with your understanding of the world into
           | insights.
        
       | GaggiX wrote:
       | Looking at the ARC problems that the model didn't solve correctly
       | I honestly have no idea in some of done on how the model of wrong
       | or what should have been the solution given the train example.
       | 
       | How strong is a human being on this challenge?
        
         | neoneye2 wrote:
         | There are only a few example images. It could be interesting
         | seeing more examples of good/bad predictions.
        
         | antognini wrote:
         | It's quite hard. You can download the dataset here [1] and it
         | comes with a little webpage so that you can try it yourself.
         | 
         | It's worth noting that you are allowed to make three guesses.
         | 
         | [1]: https://github.com/fchollet/ARC
        
       | Xcelerate wrote:
       | I always thought better sequence prediction was the natural
       | prerequisite to AGI, given that it's kind of the whole foundation
       | of algorithmic information theory (specifically, the upper limit
       | to prediction is a prefix-free universal Turing machine that
       | maximally compresses an input sequence and is allowed to continue
       | running once it has reproduced the given input sequence).
       | 
       | It's kind of funny to me that the general public seems to be
       | recognizing this in a backward fashion -- "oh hey LLMs can be
       | used for other stuff too!" I would be extremely surprised if any
       | LLM researcher hasn't heard of Solomonoff induction though.
        
       ___________________________________________________________________
       (page generated 2023-08-20 23:02 UTC)