3 ms·
AI is just good at what it's got the most elaborate training data on. And by "good" I mean, is statistically most likely to spit something out that makes some k
by Madmallard 10d ago
AI is just good at what it's got the most elaborate training data on. And by "good" I mean, is statistically most likely to spit something out that makes some kind of sense.
I don't know how well it is studied, but I suspect it is possible there is language-related complexity constraints to the effectiveness of the LLM algorithms. Like perhaps context-free grammar related problems with adequate training data can be more and more effectively solved, but maybe natural language related problems will not so much be effectively solved.
Would be curious if there is active research here.
- lelanthran 10d ago> I don't know how well it is studied, but I suspect it is possible there is language-related complexity constraints to the effectiveness of the LLM algorithms. I believe that's obvious - humans don't think in words. Neither do animals. A machine that only thinks in words is obviously going to be deficient in some things, no matter how proficient it is in everything else.
- brokensegue 10d agoAren't the new models thinking in neuralese?
- lelanthran 9d ago> Aren't the new models thinking in neuralese? Where did you read that?
- brokensegue 9d agoMultiple places
- lelanthran 9d ago> Multiple places The people you are talking to have apparently never seen this research. Maybe you can help them out and provide links? It's faster to do that than to engage in some form of audience-persuasion, as well as more effective.