Lugh@futurology.todayM to Futurology@futurology.todayEnglish · 3 months agoEvidence is growing that LLMs will never be the route to AGI. They are consuming exponentially increasing energy, to deliver only linear improvements in performance.arxiv.orgexternal-linkmessage-square68fedilinkarrow-up1297arrow-down127
arrow-up1270arrow-down1external-linkEvidence is growing that LLMs will never be the route to AGI. They are consuming exponentially increasing energy, to deliver only linear improvements in performance.arxiv.orgLugh@futurology.todayM to Futurology@futurology.todayEnglish · 3 months agomessage-square68fedilink
minus-squareGeneral_Effort@lemmy.worldlinkfedilinkEnglisharrow-up1·3 months ago In theory there’s an inflection point at which models become sophisticated enough that they can self-sustain with generating training data to recursively improve That sounds surprising. Do you have a source?
That sounds surprising. Do you have a source?