misk@sopuli.xyz to Technology@lemmy.worldEnglish · 2 days agoApple study exposes deep cracks in LLMs’ “reasoning” capabilitiesarstechnica.comexternal-linkmessage-square102fedilinkarrow-up1486arrow-down115cross-posted to: apple_enthusiast@lemmy.world
arrow-up1471arrow-down1external-linkApple study exposes deep cracks in LLMs’ “reasoning” capabilitiesarstechnica.commisk@sopuli.xyz to Technology@lemmy.worldEnglish · 2 days agomessage-square102fedilinkcross-posted to: apple_enthusiast@lemmy.world
minus-squareSemperverus@lemmy.worldlinkfedilinkEnglisharrow-up11·23 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}.
This problem is due to the fact that the AI isnt using english words internally, it’s tokenizing. There are no Rs in {35006}.