Comment on AGI achieved š¤
jsomae@lemmy.ml āØ1ā© āØmonthā© agoYouāre talking about hallucinations. Thatās different from tokenization reflection errors. Iām specifically talking about its inability to know how many of a certain type of letter are in a word that it can spell correctly. This is not a hallucination per se ā at least, itās a completely different mechanism that causes it than whatever causes other factual errors. This specific problem is due to tokenization, and thatās why I say it has little bearing on other shortcomings of LLMs.
untorquer@lemmy.world āØ1ā© āØmonthā© ago
No, Iām talking about human learning and the danger imposed by treating an imperfect tool as a reliable source of information as these companies want people to do.
Whether the erratic information is from tokenization or hallucinations is irrelevant when this is already the main source for so many people in their learning, for example, a new language.
jsomae@lemmy.ml āØ1ā© āØmonthā© ago
Hallucinations arenāt relevant to my point here. Iām not defending that AIs are a good source of information, and I agree that hallucinations are dangerous (either that or misusing LLMs is dangerous). I also admit that for language learning, artifacts caused from tokenization could be very detrimental to the user.
The point I am making is that LLMs struggling with these kind of tokenization artifacts is poor evidence for assuming anything about their behaviour on other tasks.
untorquer@lemmy.world āØ1ā© āØmonthā© ago
Thatās a fair point when these LLMs are restricted to areas where they function well. They have use cases that make sense when isolated from the ethics around training and compute. But the people who made them are applying them wildly outside these use cases.
These are pushed as a solution to every problem for the sake of profit with intentional ignorance of these issues. If a few errors impact someone itās just a casualty in the goal of making it profitable. That canāt be disentwined from them unless you limit your argument to open source local compute.
jsomae@lemmy.ml āØ1ā© āØmonthā© ago
Well ā and I donāt meant this to be antagonistic ā I agree with everything youāve said except for the last sentence where you say āand therefore youāre wrong.ā Look, Iām not saying LLMs function well, or that theyāre good for society, or anything like that. Iām saying that tokenization errors are really their own thing that are unrelated to other errors LLMs make. If you want to dunk on LLMs then yeah be my guest. Iām just saying that this one type of poor behaviour is unrelated to the other kinds of poor behaviour.