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

Scaling up test-time compute with latent reasoning: A recurrent depth approach

arxiv.org

Post reply on HN
Print view View original

41–45 of 45 posts

Re: Scaling up test-time compute with latent reasoning: A recurrent depth approach

#41

Post by lonk11 »

Tostino wrote: post #30
thomasahle wrote: post #6

Earlier quoted context omitted.

> Latent / embedding-space reasoning seems a step in the right direction Might be good for reasoning, but it's terrible for interpretation / AI-safety.

Why is it any different to do 4 recurrent passes than having a model that is 4x deeper?

Running one layer 4 times should fetch the weights of that layer once. Running 4 layers makes you fetch 4x parameters.

The recurrent approach is more efficient when memory bandwidth is the bottleneck. They talk about it in the paper.

Re: Scaling up test-time compute with latent reasoning: A recurrent depth approach

#42

Post by Tostino »

lonk11 wrote: post #41
Tostino wrote: post #30

Earlier quoted context omitted.

Why is it any different to do 4 recurrent passes than having a model that is 4x deeper?

Running one layer 4 times should fetch the weights of that layer once. Running 4 layers makes you fetch 4x parameters. The recurrent approach is more efficient when memory bandwidth is the bottleneck. They talk about it in the paper.

Yeah, understood. I'm excited for the reduction in parameter count that will come when this is taken up in major models.

I meant it rhetorically in reference to interpretability. I don't see a real difference between training a model that is 100b parameters vs a (fixed) 4x recurrent 25b parameter model as far as understanding what the model is `thinking` for the next token prediction task.

You should be able to use the same interpretability tooling for either. It can only `scheme` so much before it outputs the next token no matter if the model is just a fixed size and quite deep, or recurrent.

Re: Scaling up test-time compute with latent reasoning: A recurrent depth approach

#43

Post by bcantrill »

porridgeraisin wrote: post #28
viraptor wrote: post #24

Earlier quoted context omitted.

(citation needed) It sounds fun and all, but we barely have any connection between human brain and llms as they exist today.

We need to reboot Bryan Cantrill's "Don't anthropomorphize the lawn mower" talk with a new edition titled "Don't anthropomorphize the internet document simulator"

One step ahead of you![0]

[0] https://www.youtube.com/watch?v=bQfJi7rjuEk (slides: https://speakerdeck.com/bcantrill/intelligence-is-not-enough...)

Re: Scaling up test-time compute with latent reasoning: A recurrent depth approach

#44

Post by ckrapu »

optimalsolver wrote: post #15
ckrapu wrote: post #9

My opinion is that opaque reasoning is a prerequisite for many of the worst possible AI outcomes. We should make reasoning fully visible in the output space.

Is there any actual evidence that the reasoning tokens output by current models actually represent the computation happening in the hidden layers? In both cases, the model is doing a ton of processing that you can't actually inspect, except here, you at least get some efficiency gains. Even more importantly, you're also less likely to convince yourself that you know what the model is thinking.

In the autoregressive decoding framework, the hidden layers' state for computation of token `t` is conditionally independent of all hidden states for `t-1`, `t-2` and so on given the observed tokens.

Put differently, the observed tokens are a bottleneck on the information that can be communicated across tokens. Any scheming performed by an LLM which requires more than one token to formulate must therefore pass through the visible tokens. With opaque vectors transferred across decoding steps, this is not the case.

The computation in the hidden layers, as far as we can tell, is not sufficient for scheming in a single decoding step. It looks like it requires O(10^2) or O(10^3) steps instead, judging from anecdotal evidence like the reports of scheming from o1 (https://cdn.openai.com/o1-system-card-20241205.pdf)

As far as your last point goes, I'd rather have a more transparent system, all other factors held constant.

Re: Scaling up test-time compute with latent reasoning: A recurrent depth approach

#45

Post by porridgeraisin »

bcantrill wrote: post #43
porridgeraisin wrote: post #28

Earlier quoted context omitted.

We need to reboot Bryan Cantrill's "Don't anthropomorphize the lawn mower" talk with a new edition titled "Don't anthropomorphize the internet document simulator"

One step ahead of you![0] [0] https://www.youtube.com/watch?v=bQfJi7rjuEk (slides: https://speakerdeck.com/bcantrill/intelligence-is-not-enough... )

Nice, right from the horse's mouth. Let me watch that.
Post reply on HN