misk to Technology@lemmy.worldEnglish · 6 months agoApple study exposes deep cracks in LLMs’ “reasoning” capabilitiesarstechnica.comexternal-linkmessage-square109linkfedilinkarrow-up1508arrow-down119cross-posted to: apple_enthusiast@lemmy.worldarstechnica_index@rss.ponder.cat
arrow-up1489arrow-down1external-linkApple study exposes deep cracks in LLMs’ “reasoning” capabilitiesarstechnica.commisk to Technology@lemmy.worldEnglish · 6 months agomessage-square109linkfedilinkcross-posted to: apple_enthusiast@lemmy.worldarstechnica_index@rss.ponder.cat
minus-squareSemperverus@lemmy.worldlinkfedilinkEnglisharrow-up12arrow-down1·6 months 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}.