The Picard Maneuver@lemmy.world to People Twitter@sh.itjust.works · 1 day agoIt's best not to dwell on itlemmy.worldimagemessage-square129fedilinkarrow-up11.64Karrow-down13
arrow-up11.64Karrow-down1imageIt's best not to dwell on itlemmy.worldThe Picard Maneuver@lemmy.world to People Twitter@sh.itjust.works · 1 day agomessage-square129fedilink
minus-squaremelpomenesclevage@lemmy.dbzer0.comlinkfedilinkEnglisharrow-up15·1 day agothat depends on what topic you know and how well you know it.
minus-squaretaladar@sh.itjust.workslinkfedilinkarrow-up11·edit-26 hours agoLLMs are actually pretty good for looking up words by their definition. But that is just about the only topic I can think of where they are correct even close to 80% of the time.
minus-squaremelpomenesclevage@lemmy.dbzer0.comlinkfedilinkEnglisharrow-up1·1 day agoyeah. some things I’d be shocked if they were correct 1% of the time. some things, like that, I might expect them to be correct about 80%, yeah.
40% seems low
that depends on what topic you know and how well you know it.
LLMs are actually pretty good for looking up words by their definition. But that is just about the only topic I can think of where they are correct even close to 80% of the time.
yeah. some things I’d be shocked if they were correct 1% of the time. some things, like that, I might expect them to be correct about 80%, yeah.