alt Hacker NewsIt's still clear that LLMs lack spatial reasoning, either in the concrete or abstract, and while that sort of reasoning has been downplayed by academia for at least a century it is fundamental to technology and industry. (And many would say for science and mathematics too).
They will, however, get there as well either directly or as interfaces to models that do, and your core point stands.
Replies
simianwords • today at 2:42 PM
Is there any proof that they are not good at special reasoning? Arc agi 1 and 2 are saturated.
➕ show 3 replies
"Lack" isn't the right word. "Lacking" is more like it.
If there was a deep fundamental inability, we wouldn't see things like newer generations of LLMs consistently improving on ARC-AGI series (heavy spatial reasoning loading) and SimpleBench (a lot of commonsense + spatial reasoning components).
In a way, it's a surprise that LLMs, notoriously lacking any sort of embodied experience, can even get this close to human baselines on tasks like this.
My takeaway is that text is a far richer modality than anyone has expected - and that high end LLMs are often sharp and flexible enough to recognize their weak points and substitute their strengths. I.e. all the LLMs implementing A* to optimally solve pathfinding in ARC-AGI-3 tasks, often unprompted.
There might still be unrealized gains there from true depth-unbounded recurrence, or maybe from finding better ways to integrate modalities in training. But clearly, a "fundamental limit" it ain't.