No, you cannot.
I rarely do anything with LMs [1], but if I do I ask them to code "blocked clause decomposition" [2] in Haskell. Blocked clause decomposition is very, very simple as far as algorithm's implementation complexity goes and Haskell is just an unusual but simple implementation language.
[1] https://arxiv.org/pdf/2403.07183.pdf
[2] https://www.cs.utexas.edu/~marijn/publications/bcd.pdf
The Phind's responces were very illuminating ones.
On a first try, it tried to make me code that myself by pointing to a paper that does not contain any blocked clause decomposition algorithm(s), just worthless combinations of them and references to other papers. Of course, I read that paper before. When I specifically asked to provide me with code, it did not even provided me with the correct type of blocked clause, which should be a pair of a CNF clause and a literal that blocks it.
On a second try I could not even pass through bullshitting phase - I just was not able to persuade a LM to write any code. The second try I attempted after I saw an advert here on HN about Phind's new and updated LM, better at coding.
Try that youself with your unusual but known problem of choice.
And this is exactly a problem of collapse of knowledge: you can't ask an LM to code you a problem it hasn't seen several dozen of times during training. You just cannot, you will not get a satisfying answer.
I think that's more a function of the coding ability of the model you are using, and obviously they are going to do much better on languages better represented in the training set.
What the paper is referring to as "knowledge collapse" isn't inability to do something, but rather getting a mid response since by definition that's going to dominate any normal (pardon the pun!) distribution.
Here's what he said:
"While large language models are trained on vast amounts of diverse data, they naturally generate output towards the 'center' of the distribution."
But, you can generally prompt around this sort of thing by asking the model to adopt a preferred context such as "you're an expert xxx", providing of course that is did see such expert-tagged material in the training set.
So, I need LM to be an expert in SAT solving and an expert Haskell programmer. This means I need LM to be trained on the SAT solving and on vast amount of Haskell code.
I think it is already easy to see that LM should be trained in an intersection of quite fringe (yes, I am watching that TV series right now) things.
I also think that it is beyond reason to ask for this.
I recommend refraining from making (and especially from posting) any conclusions until then.
Strongly reminds me of "no true scotsman."
If you have an access to a good model, try that simple problem of blocked clause decomposition and please get back to us.
Please preface your future posts on the topic with “I’ve never tried any decent models, so I don’t know what I’m talking about”.
I, actually, asked my friend to ask whatever LLM he paid for to perform this experiment. I think he uses latest or previous GPT. At least he is paying for it which I cannot do and won't do in foreseeable future.
Just like I said above, not only LLM failed to properly implement the algorithm, it failed to provide even remotely correct types for clauses and blocked clauses.
What happened next amused me like nothing else related to AI.
The exchange that followed incorrect initial attempt, GPT clearly tried to persuade my friend to implement the correct version for it, constantly providing him with partially (and stupidly) incorrect code.
This is exactly what MathCraft [1] does.
[1] https://en.wikipedia.org/wiki/Cyc#MathCraft
And this behavior is clealy counterproductive. This is not what one might expect from anything that is advertized as a good assisting measure. I can use ghci, vim and actual paper with much greater effect.
Essentially why the current batch of top-tier LLMs are trained on a pile of text amounting to “the internet”, containing every haskell q&a on stackoverflow, every sat implementation on github, and every explaination of those terms on wikipedia. It might seem unreasonable without the sense of scale and expense of building these things; but yes, it has those input data in its training set.
For LLM to implement BCD in Haskell it needs to be able to translate between Haskell and C++ and it is well beyond trivial.
Which, again, indicates that the current crop of LLM’s are just fuzzy-recollection/fuzzy-interpolation machines, and have none of the actual thought/reasoning capabilities some people are claiming they have.
If we think in terms of automated plagiarism, no one can complain that the model doesn't produce original ideas.
Automated plagiarism could have many uses but it is obviously stupid to expect an original idea from plagiarism. We wouldn't complain the model can't reproduce things it hasn't seen before if we think in terms of plagiarism.
Basically Chompsky has been proved wrong by LLMs, but rather than believing what is in front of his eyes he is saying "that can't be right, because it'd mean I am wrong".
Mmmhhmmm yes very convincing argument there. I’d be willing to bet that Chomsky’s arguments don’t rely on such spurious arguments. Automated plagiarism reads like a fantastic description of current LLM’s to me.
Since they do at least (whatever other shortcomings they may have) appear to have learnt language, just by "listening to" a bunch of examples, and since they are based on a transformer, not some specialized Chompskian "language organ", then the thesis that a "language organ" is needed is disproved.
If you look at the cortex as a whole, what seems most striking is how uniform it is - same basic architecture of six layers of neurons with a specific pattern of inter-layer connectivity. It seems nature has come up with a universal prediction architecture. Maybe Wernicke's areas is fine-tuned for language, but to characterize it as a specialized "language organ" seems a bit of a stretch. Let's note too that language is only a million years old, while the cortex itself is 100's of millions, yet has this mostly uniform architecture that evidentially works just as well for vision as hearing, etc, etc.
So, sure, the ability of the ridiculously simple transformer architecture to learn language (and many other prediction tasks you throw at it), doesn't PROVE that the brain didn't have to evolve a highly specialized way to do it, but it certainly seems highly suggestive of it.
Since we now have an existence proof that a very simple architecture, not specialized for language, can learn language, it seems the onus is now on Chompsky to put some meat on the bones of his claim (without evidence) that the general cortical architecture is incapable of this without a high degree of specialization.
On the contrary, it's extremely likely that we can engineer something better than random evolution. And we have: calculators are better at adding and subtracting that humans are, bikes are faster than running, etc.
> If you look at the cortex as a whole, what seems most striking is how uniform it is - same basic architecture of six layers of neurons with a specific pattern of inter-layer connectivity. It seems nature has come up with a universal prediction architecture. Maybe Wernicke's areas is fine-tuned for language, but to characterize it as a specialized "language organ" seems a bit of a stretch. Let's note too that language is only a million years old, while the cortex itself is 100's of millions, yet has this mostly uniform architecture that evidentially works just as well for vision as hearing, etc, etc.
Language is ~100,000 years old, not a million. The brain is in fact not uniform: If you damage Broca's area you lose language capability. And to say that all cognitive function is the same general algorithm ignores the obvious fact that the brain performs different functions and doesn't answer the question of how and why this is so. There are lots of cognitive behaviour that is not understood, if you want to explain those behaviors you implicitly have to distinguish them from one another.
> So, sure, the ability of the ridiculously simple transformer architecture to learn language (and many other prediction tasks you throw at it), doesn't PROVE that the brain didn't have to evolve a highly specialized way to do it, but it certainly seems highly suggestive of it.
It doesn't. Moro showed in a series of experiments that humans have difficulty learning non-hierarchical languages and use different parts of the brain to do so, which is highly suggestive that language is specialized.
> Since we now have an existence proof that a very simple architecture, not specialized for language, can learn language, it seems the onus is now on Chompsky to put some meat on the bones of his claim (without evidence) that the general cortical architecture is incapable of this without a high degree of specialization.
He has provided evidence and arguments, some of which I have pointed to above. Maybe you should actually read or listen to him.
The cortex is built for hierarchical processing, because that's what's needed to model the world we live in. Physical objects are localized and have hierarchical detail, and larger visual scenes are the same. The kind of sequential (temporal) patterns relevant to us are also hierarchical, whether in visual, auditory or other domains.
The type of connectionist architecture needed to recognize hierarchical patterns is a layered one where the receptive field and hierarchical level of abstraction grows as you ascend the layers. In our brain those layers come from different patches of our cortical sheet being connected. This is the reason that neural network architectures like CNNs and transformers also work to recognize hierarchical patterns in visual and temporal domains - because they both also uses these layered architectures, which is all that is needed.
The reason why the function of damaged brain areas can't always be taken over by other areas is largely due to plasticity. Our brain peaks in it's ability to form new synapses in the first few years of life. If you haven't learned language by age of 3, then you will never be able to learn more than a crude type of pidgin language, depsite all your "language areas" being intact. The same would be true of different part of you brain trying to learn language as an adult - the plasticity is no longer there.
Learning a programming language (or any complex skill) just by reading a book and looking at examples, without actually practicing it yourself, is basically impossible, even for ourselves, let alone an LLM. It is amazing how well the LLMs can do at programming in cases where they were trained on masses of well-commented examples, but without also being able to learn "on the job" themselves, you need to tamper expectations.
Domain expert: “I cannot get it to produce something that even approximately resembles x”
Person: “I gave it the paper and I got maybe x?”
There is a fundamental knowledge gap here, just because you can get it to do something isn’t the goal, or even a good outcome. The illusion that it’s capable of doing something it evidently cannot is a dangerous one, because people who don’t know better, and are being convinced that they do understand or comprehend when they don’t.
If you are just using the free version of Phind, then it uses some MUCH smaller and less capable model of their own. Even if you use Phind Pro, or Phind Plus, it says their own most capable model is Phind 70B (tiny compared to GPT-4 class models).
You can't just use Phind and take that as indicative of what state-of-the-art models can do, without knowing what model you are actually using. You also can't expect any of these models to do as well with Haskell as they do with Python or Java which they will have seen massively more of during training.
This is exactly the person’s point.
Did you try this Haskell problem with a paid version and can provide an example that actually disputes the described experience? Why would someone pay for the premium version of a product when they don’t perceive the free version as working?
In keyword-based indexing solutions, a document vector is created using "term frequency inverse document frequency" scores. The idea is to pump up the document on the dimension where the document is unique compared to the other documents in the corpus. So when a query is issued with emphasis on a certain dimension, only documents that has higher scores in that dimension are returned.
But the uniqueness in those solutions is based on keywords being used in the document, not concepts.
What we need here to eliminate "blandness" is conceptual uniqueness. Maybe TF-IDF is still relevant to get there. Something to think about.
The model hasn't forgotten the diversity of material it was trained on, but outside of a context predicting a "long tail" response, it's going to predict a mid response. You can always prompt it to respond differently though.
Blandness is more of an issue since that's what most-probable word-by-word generation is going to give you, rather than the less predictable, but more interesting, responses that an individual might give. Prompting could help by asking the model to reply in the idiosyncratic style of some celebrity, but this is likely to come across as a cheesy impression. Maybe the models could be trained to generate conditioned on a provided style sample, which could be long enough to avoid the cheesiness.