David GerardM to [email protected]English • 2 months agoRemember how ChatGPT totally aced the bar exam? Wow! yeah, turns out that was just a liewww.nytimes.commessage-square252fedilinkarrow-up1648file-textcross-posted to: [email protected]
arrow-up1648external-linkRemember how ChatGPT totally aced the bar exam? Wow! yeah, turns out that was just a liewww.nytimes.comDavid GerardM to [email protected]English • 2 months agomessage-square252fedilinkfile-textcross-posted to: [email protected]
minus-square@[email protected]linkfedilinkEnglish8•2 months agoeven if that wasn’t the case, a 90% success rate is absolutely abysmal in practice.
minus-square@[email protected]linkfedilinkEnglish46•2 months ago90th percentile means it performed equal or better than 90% of the comparisons, no? Not that it got 90% score.
even if that wasn’t the case, a 90% success rate is absolutely abysmal in practice.
90th percentile means it performed equal or better than 90% of the comparisons, no? Not that it got 90% score.