Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Scaling Transformer to 1M tokens and beyond with RMT

arxiv.org

Post reply on HN
Print view View original

31–40 of 147 posts

Re: Scaling Transformer to 1M tokens and beyond with RMT

#31

Post by nl »

kgwgk wrote: post #29
nl wrote: post #26

For those who don't completely get the impact of this: We already know Large Language Models (LLMs) can learn at runtime (ie, separately to the training process.) This is called "In Context Learning". See [1], [2] for more details. (BTW, when anyone says "LLMs are stochastic parrots" you know they are ignorant of this) In context learning is wonderful because it means you can "train" a LLM at run time by filling the…

> (BTW, when anyone says "LLMs are stochastic parrots" you know they are ignorant of this) Do you? Some of them -at least- understand that the output is conditional on the input.

"Conditioned" with the ability to introduce new facts and have the model infer impacts on them.

As an example, tell chatGPT that the Queen of England died (which occurred after the data cut off) and then ask it who the head of state of Australia is.

It's able to infer the head of state of Australia is now Charles III (and gives a good explanation of how this is mostly ceremonial.) See https://twitter.com/nlothian/status/1646699207506685953

At some point the word "stochastic" doesn't really capture that behavior in any useful sense.

Re: Scaling Transformer to 1M tokens and beyond with RMT

#32

Post by tomohelix »

kgwgk wrote: post #29
nl wrote: post #26

For those who don't completely get the impact of this: We already know Large Language Models (LLMs) can learn at runtime (ie, separately to the training process.) This is called "In Context Learning". See [1], [2] for more details. (BTW, when anyone says "LLMs are stochastic parrots" you know they are ignorant of this) In context learning is wonderful because it means you can "train" a LLM at run time by filling the…

> (BTW, when anyone says "LLMs are stochastic parrots" you know they are ignorant of this) Do you? Some of them -at least- understand that the output is conditional on the input.

>the output is conditional on the input

Uh, isn't that how it should always be? If the output isn't conditional on the input, then it is basically noise or worthless. How the conditionals are setup are the key but not really sure how does this relate to the point you are making.

Re: Scaling Transformer to 1M tokens and beyond with RMT

#33

Post by ad404b8a372f2b9 »

zwaps wrote: post #28

I really wish that CS would switch to discussing related literature before (or with) contributions. The authors only cite themselves until the very end of the paper, and the whole time you are thinking: "Hasn't this been done before? Transformer-XL (recurrent Memory) is a years old paper!" And you need to read to the very and to find out that the contribution is actually: Keep your architecture (but still need to tra…

I feel like it used to be much more common to have "Related Works" as part 2 or way earlier in papers in earlier years.

Re: Scaling Transformer to 1M tokens and beyond with RMT

#34

Post by albertzeyer »

This looks like a potential NeurIPS submission.

But it will probably be rejected. The quality bar for NeurIPS is quite high.

Some reasons:

The experiments are very weak: There are just a few figures, basically figure 1 and figure 5, which show some results. There are no tables with numbers.

But more importantly: There are no comparisons (in terms of experiments/numbers) to similar models, like:

- Block-Recurrent Transformers (https://arxiv.org/abs/2203.07852) and related approaches to make the Transformer recurrent, so effectively getting infinite context length.

- All the work on sparse attention, or linear attention, like Longformer, etc, which should also allow for such context lengths.

I don't mean that they just mention them in related work (they partly do, although it looks very short, so I'm quite sure they leave out a lot of other related work, without looking too much into it). I mean that they really run experiments and show numbers. And also, to look at the model definitions of those alternatives, and compare and analyze the differences.

Analysis also seems to be a bit weak. Not sure about novelty.

So, while the presented approach could be interesting, it's hard to tell how good it really performs, and how it compares to alternatives.

(This is now a 10 min review by me. If I had to review this, I usually would spend more like 1-2h at least, so giving much more details. But I don't expect that my initial impression would change too much.)

Re: Scaling Transformer to 1M tokens and beyond with RMT

#35

Post by nl »

macrolime wrote: post #30
nl wrote: post #23

Earlier quoted context omitted.

No reason why Toolformer+GPT couldn't use a symbolic reasoning program. Symbolic reasoning programs are much better than humans. Probably bugs will happen in the interface between human language and the symbolic reasoning. But that happens with humans too.

GPT-4 has toolformer capabilities as emergent properties, so using it in combination with toolformer is really unnecessary.

This is true too. Even chatGPT can learn some tool use in-context. See https://twitter.com/minosvasilias/status/1627076214639976449

Re: Scaling Transformer to 1M tokens and beyond with RMT

#36

Post by kgwgk »

tomohelix wrote: post #32
kgwgk wrote: post #29

Earlier quoted context omitted.

> (BTW, when anyone says "LLMs are stochastic parrots" you know they are ignorant of this) Do you? Some of them -at least- understand that the output is conditional on the input.

>the output is conditional on the input Uh, isn't that how it should always be? If the output isn't conditional on the input, then it is basically noise or worthless. How the conditionals are setup are the key but not really sure how does this relate to the point you are making.

According to the parent comment anyone who says "LLMs are stochastic parrots" is ignorant of the effect that changing the input “filling the context with examples” has on the output.

Re: Scaling Transformer to 1M tokens and beyond with RMT

#37

Post by geraneum »

nl wrote: post #26

For those who don't completely get the impact of this: We already know Large Language Models (LLMs) can learn at runtime (ie, separately to the training process.) This is called "In Context Learning". See [1], [2] for more details. (BTW, when anyone says "LLMs are stochastic parrots" you know they are ignorant of this) In context learning is wonderful because it means you can "train" a LLM at run time by filling the…

> In context learning is wonderful because it means you can "train" a LLM at run time

The “learning” in “In context learning” is not the same concept as learning during the network “training”. The former is loosely named and does not have any impact on what’s actually “learned” by the model. The LLM can, by nature, contradict its training data and whatever you/it prompt(s), and incorporates randomness when completing your prompts, hence the term “stochastic”.

Nevertheless, it’s useful to have bigger context window if it works well.

Re: Scaling Transformer to 1M tokens and beyond with RMT

#38

Post by aezart »

nl wrote: post #22
alchemist1e9 wrote: post #12

Earlier quoted context omitted.

Do we yet have a single example of a transformer based AI, i.e. LLM, learning something new that we didn’t teach it in the training data? Maybe we do. I’m not sure how to define it, but new should be a discovery or insight or even relationship that is not explicitly taught in the training data. If we don’t then it suggests without humans in the loop that the super intelligence is not so close?

There are plenty of examples of "new" things a LLM can do. A good example is all those toy examples of "Program a whatever in the style of Shakespeare and David Bowie's love child". This isn't a thing that it has seen in training data.

From my limited understanding of how LLMs work, I believe this behavior is enabled by embeddings. The model maps its ~50,000 token vocabulary into a lower-dimensional vector space. Each dimension in the vector adds some sort of meaning (or at least association) to the word.

I saw an example in a Numberphile video where they were able to take the vector of the word "prince", subtract the vector of the word "man", add the vector of the word "woman", and the resulting vector was closest to the word "princess". So in theory there could be a "gender" dimension and a "position of authority" dimension in that vector (or the model might be making other, stranger connections between words that we don't understand).

I think the same thing is happening in your example. The model identifies and produces output that keeps it in the general region in the vector space for the "Shakespearean" and "Bowiean" dimensions while still satisfying other requirements.

Re: Scaling Transformer to 1M tokens and beyond with RMT

#40

Post by nl »

kgwgk wrote: post #36
tomohelix wrote: post #32

Earlier quoted context omitted.

>the output is conditional on the input Uh, isn't that how it should always be? If the output isn't conditional on the input, then it is basically noise or worthless. How the conditionals are setup are the key but not really sure how does this relate to the point you are making.

According to the parent comment anyone who says "LLMs are stochastic parrots" is ignorant of the effect that changing the input “filling the context with examples” has on the output.

That isn't a fair summary of what I'm saying.

People who say "LLMs are stochastic parrots" may well be aware of effects of conditioning the input.

This in itself does not completely describe the capabilities of a LLM, since their ability to learn and use those new facts to override their previous "beliefs" is not what you'd expect from mere stochastic behavior.

My point is that "stochastic parrot" believers are unaware of this.

Post reply on HN