LLMs do not reason. Not hard to understand.
There have been enough cases of models providing novel results that it's clear that whatever human trait they supposedly lack they don't really need. A car does not need legs, it does things differently. Having legs would even be a major detriment and would hold it back from achieving its top performance.
That's what those brain simulating projects are conceptually btw: cars with legs or planes with flapping wings. That's why they all fail, the approach makes no sense.
Those are some mighty parrots there, if they managed to get gold at IMO, IoI, and so on...
>They are a parrot
Is it really much different from most people? The average Joe doesn't produce novel theories every day - he just rehashes what he's heard. Now the new goalpost seems to be that we can only say an LLM can "reason" if it matches Fields Medalists.
You've presented a false choice.
However the average Joe does indeed produce unique and novel thoughts every day. If it were not the case he would be brain dead. Each decision - wearing blue or red today - every tiny thought, action, feeling, indecision, crisis, or change of heart these are just as important.
The jury maybe out on how to judge what 'thought' actually is. However what it is not is perhaps easier to perceive. My digital thermometer does not think when it tells me the temperature.
My paper and pen version of the latest LLM (quite a large bit of paper and certainly a lot of ink I might add) also does not think.
I am surprised so many in the HN community have so quickly taken to assuming as fact that LLM's think or reason. Even anthropomorphising LLM's to this end.
For a group inclined to quickly calling out 'God of the gaps' they have quite quickly invented their very own 'emergence'.
Even if we're to humor the "novel" part, have they actually come up with anything truly novel? New physics? New proofs of hard math problems that didn't exist before?
[0] https://research.google/blog/ai-as-a-research-partner-advanc...