Imagine I want to attend a conference in a different country. Google maps might give turn by turn navigation but that is an overwhelming and largely irrelevant mess of details for most planning purposes. Eg: all I might want to know is the different flight legs and the fact that the journey takes 15-18 hours, and not all the turns and traffic lights to get from home to the airport.
I want a zoomed out picture, and to be able to fill in detail hierarchically, on demand.
Instead, one-step models give you the full high-res local structure of the graph that would have to search through (with too many states and edges).
This is the same reasoning behind why Yann Lecun thought test-time scaling would not work for LLMs: compounding error.
Instead, the more tokens LLMs use, the better their performance on many tasks. LLMs can self-correct, evidenced by the power of getting models to question themselves by emitting "Wait," in S1. https://arxiv.org/abs/2501.19393
You wouldn't believe the amount of reasoning I saw these past few months that was correct until the stochastic parrot decided that a "wait" token should now be used and everything steered off a cliff.
Yeah came here to comment exactly this. And this is generally why I dislike/avoid this type of first principle analysis: it can make very convincing arguments that are just totally wrong due to some misleading assumption
I'm not sure I follow what one step means exactly. Aren't all models some f(x) = y? Is the suggestion instead that we should be doing f(x) = g(h(x)) = y?
12 comments
[ 3.1 ms ] story [ 31.2 ms ] threadI want a zoomed out picture, and to be able to fill in detail hierarchically, on demand. Instead, one-step models give you the full high-res local structure of the graph that would have to search through (with too many states and edges).
Instead, the more tokens LLMs use, the better their performance on many tasks. LLMs can self-correct, evidenced by the power of getting models to question themselves by emitting "Wait," in S1. https://arxiv.org/abs/2501.19393
What would the difference be?