5 ms·
The simpler and I think correct conclusion is that the LLM simply does not reason in our sense of the word. It mimics the reasoning pattern and try to get it ri
by zadwang 8mo ago
The simpler and I think correct conclusion is that the LLM simply does not reason in our sense of the word. It mimics the reasoning pattern and try to get it right but could not.
- esafak 8mo agoWhat do you make of human failures to reason then?
- dns_snek 8mo agoHumans who fail to reason correctly with similar frequency aren't good at solving that task, same as LLMs. For the N-th time, "LLM is as good at this task as a human who's bad at it" isn't a good selling point.
- esafak 8mo agoYou didn't claim that such humans fail to "reason in our sense of the word". Why are you not holding them up to the same standard?
- dns_snek 8mo agoI didn't but I'm happy to make that claim - humans who exhibit that sort of behavior aren't reasoning either, and they are just as unpleasant to deal with as LLMs are.
- cindyllm 8mo ago[dead]