3 ms·

But what is novel content? I can easily falsify the accusation that, "people underestimate transformers and don't see that they are actually intelligent," by d

But what is novel content? I can easily falsify the accusation that, "people underestimate transformers and don't see that they are actually intelligent," by defeating the best open-source transformer-based word embedding (at the time) with a simple TF-DF based detector (this was back in September). https://www.patdel.com/plagiarism-detector/ https://www.patdel.com/plagiarism-detector/ No, these things are not, "emergent," they are just rearranging numbers. You don't have to use a transformer or neural network at all to re-arrange numbers and create something that is even more, "artificially intelligent," than one that does use transformers it turns out!

whimsicalism 4y ago
I can tell that this conversation is not going to be super productive, so a few brief thoughts: > I can easily falsify the accusation that, "people underestimate transformers and don't see that they are actually intelligent," I think that you have an idiosyncratic definition of what "falsify" means compared to what most might. Getting away from messy definitions of "intelligent" which I think are value-laden, I see nothing in your blog post that falsifies the notion that LLMs can generate novel content (another fuzzy value-laden notion perhaps). > these things are not, "emergent," they are just rearranging numbers. It seems non-obvious to me that 'rearranging numbers' cannot lead to anything emergent out of that process, yet cascading voltage (as in our brain) can.
- wwwpatdelcom 4y ago
  I would love to read anything you have written or studied about this topic at length. Thanks for your replies.
hackinthebochs 4y ago
>No, these things are not, "emergent," they are just rearranging numbers. This is a bad take. Most ways to "rearrange numbers" produce noise. That there is a very small subset of permutations that produce meaningful content, and the system consistently produces such permutations, is a substantial result. The question of novelty is whether these particular permutations have been seen before, or perhaps are simple interpolations of what has been seen before. I think its pretty obvious the space of possible meaningful permutations is much larger than what is present in the training set. The question of novelty then is whether the model can produce meaningful output (i.e. grammatically correct, sensible, plausible) in a space that far outpaces what was present in the training corpus. I strongly suspect the answer is yes, but this is ultimately an empirical question.
- wwwpatdelcom 4y ago
  I would love to read anything you have written about the topic at length. Thanks for your contribution.
  - hackinthebochs 4y ago
    I haven't written anything substantial on the subject unfortunately. I do have some ideas rattling around so maybe this will motivate me to get them down.

whimsicalism 4y ago

I can tell that this conversation is not going to be super productive, so a few brief thoughts: > I can easily falsify the accusation that, "people underestimate transformers and don't see that they are actually intelligent," I think that you have an idiosyncratic definition of what "falsify" means compared to what most might. Getting away from messy definitions of "intelligent" which I think are value-laden, I see nothing in your blog post that falsifies the notion that LLMs can generate novel content (another fuzzy value-laden notion perhaps). > these things are not, "emergent," they are just rearranging numbers. It seems non-obvious to me that 'rearranging numbers' cannot lead to anything emergent out of that process, yet cascading voltage (as in our brain) can.

wwwpatdelcom 4y ago
I would love to read anything you have written or studied about this topic at length. Thanks for your replies.

hackinthebochs 4y ago

>No, these things are not, "emergent," they are just rearranging numbers. This is a bad take. Most ways to "rearrange numbers" produce noise. That there is a very small subset of permutations that produce meaningful content, and the system consistently produces such permutations, is a substantial result. The question of novelty is whether these particular permutations have been seen before, or perhaps are simple interpolations of what has been seen before. I think its pretty obvious the space of possible meaningful permutations is much larger than what is present in the training set. The question of novelty then is whether the model can produce meaningful output (i.e. grammatically correct, sensible, plausible) in a space that far outpaces what was present in the training corpus. I strongly suspect the answer is yes, but this is ultimately an empirical question.

wwwpatdelcom 4y ago
I would love to read anything you have written about the topic at length. Thanks for your contribution.
- hackinthebochs 4y ago
  I haven't written anything substantial on the subject unfortunately. I do have some ideas rattling around so maybe this will motivate me to get them down.