misk@sopuli.xyz to Technology@lemmy.worldEnglish · 1 day agoApple study exposes deep cracks in LLMs’ “reasoning” capabilitiesarstechnica.comexternal-linkmessage-square94fedilinkarrow-up1468arrow-down114
arrow-up1454arrow-down1external-linkApple study exposes deep cracks in LLMs’ “reasoning” capabilitiesarstechnica.commisk@sopuli.xyz to Technology@lemmy.worldEnglish · 1 day agomessage-square94fedilink
minus-squareJDPoZ@lemmy.worldlinkfedilinkEnglisharrow-up11·20 hours ago …a spellchecker on steroids. Ask literally any of the LLM chat bots out there still using any headless GPT instances from 2023 how many Rs there are in “strawberry,” and enjoy. 🍓
minus-squareSterile_Technique@lemmy.worldlinkfedilinkEnglisharrow-up4·9 hours agoThat was both hilarious and painful. And I don’t mean to always hate on it - the tech is useful in some contexts, I just can’t stand that we call it ‘intelligence’.
minus-squareSemperverus@lemmy.worldlinkfedilinkEnglisharrow-up11·16 hours agoThis problem is due to the fact that the AI isnt using english words internally, it’s tokenizing. There are no Rs in {35006}.
Ask literally any of the LLM chat bots out there still using any headless GPT instances from 2023 how many Rs there are in “strawberry,” and enjoy. 🍓
That was both hilarious and painful.
And I don’t mean to always hate on it - the tech is useful in some contexts, I just can’t stand that we call it ‘intelligence’.
This problem is due to the fact that the AI isnt using english words internally, it’s tokenizing. There are no Rs in {35006}.