☆ Yσɠƚԋσʂ ☆ to [email protected]English • 2 years agoChatGPT gets code questions wrong 52% of the timewww.theregister.comexternal-linkmessage-square66fedilinkarrow-up1207cross-posted to: [email protected]
arrow-up1207external-linkChatGPT gets code questions wrong 52% of the timewww.theregister.com☆ Yσɠƚԋσʂ ☆ to [email protected]English • 2 years agomessage-square66fedilinkcross-posted to: [email protected]
minus-squareFuckass [none/use name]linkfedilinkEnglish19•2 years agoIt’s incredible how LLM started off as an almost miraculous software that generated impressive answers but now it’s just House Server of Leaves
minus-square@[email protected]linkfedilinkEnglish6•2 years agoThe trick is you have to correct for the hallucinations, and teach it to revert back to a health path when going off course. This isn’t possible with current consumer tools.
minus-squareMarxism-FennekinismlinkfedilinkEnglish10•2 years agoBecause it was marketing hype (read: marketing propaganda).
It’s incredible how LLM started off as an almost miraculous software that generated impressive answers but now it’s just
HouseServer of LeavesThe trick is you have to correct for the hallucinations, and teach it to revert back to a health path when going off course. This isn’t possible with current consumer tools.
Because it was marketing hype (read: marketing propaganda).