▲ 29 ▼ Large Language Model Performance Doubles Every 7 Months (spectrum.ieee.org) submitted 1 year ago by flango@lemmy.eco.br to c/technology@lemmy.world 54 comments fedilink hide all child comments By 2030, AI will greatly outperform humans in some complex intellectual tasks. Discover how LLMs are doubling their capabilities every seven months.
[–] goondaba@lemmy.world 85 points 1 year ago (2 children) *with 50 percent reliability. Heck of an asterisk on this claim. permalink fedilink source hideshow 4 child comments replies: [–] eager_eagle@lemmy.world 15 points 1 year ago (2 children) That sounds like a coin flip, but 50% reliability can be really useful. If a model has 50% chance of completing a task that would cost me an hour - and I can easily check it was completed correctly - on average, I'm saving half of the time it would take to complete this. That said, exponentials don't exist in the real world, we're just seeing the middle of a sigmoid curve, which will soon yield diminishing returns. permalink fedilink source parent hideshow 4 child comments replies: [–] prole@lemmy.blahaj.zone 6 points 1 year ago and I can easily check it was completed correctly Can you always though? permalink fedilink source parent [–] Womble@lemmy.world 1 point 1 year ago (1 child) That said, exponentials don’t exist in the real world, we’re just seeing the middle of a sigmoid curve, which will soon yield diminishing returns. Yes, but the tricky thing is we have no idea when the seemingly exponential growth will flip over into the plateuing phase. We could be there already or it could be another 30 years. For comparison Moores law is almost certainly a sigmoid too, but weve been seeing exponential growth for 50 years now. permalink fedilink source parent hideshow 2 child comments replies: [–] squaresinger@lemmy.world 3 points 1 year ago (1 child) Moore's law hasn't been exponential for ~15 years now. permalink fedilink source parent hideshow 2 child comments replies: [–] Womble@lemmy.world 1 point 1 year ago If you are just talking transitor density I believe it still is, but even if not, my point was that it had exponential growth spanning over many decades. permalink fedilink source parent [–] spankmonkey@lemmy.world 1 point 1 year ago All that power used for a fucking coin flip. permalink fedilink source parent
[–] eager_eagle@lemmy.world 15 points 1 year ago (2 children) That sounds like a coin flip, but 50% reliability can be really useful. If a model has 50% chance of completing a task that would cost me an hour - and I can easily check it was completed correctly - on average, I'm saving half of the time it would take to complete this. That said, exponentials don't exist in the real world, we're just seeing the middle of a sigmoid curve, which will soon yield diminishing returns. permalink fedilink source parent hideshow 4 child comments replies: [–] prole@lemmy.blahaj.zone 6 points 1 year ago and I can easily check it was completed correctly Can you always though? permalink fedilink source parent [–] Womble@lemmy.world 1 point 1 year ago (1 child) That said, exponentials don’t exist in the real world, we’re just seeing the middle of a sigmoid curve, which will soon yield diminishing returns. Yes, but the tricky thing is we have no idea when the seemingly exponential growth will flip over into the plateuing phase. We could be there already or it could be another 30 years. For comparison Moores law is almost certainly a sigmoid too, but weve been seeing exponential growth for 50 years now. permalink fedilink source parent hideshow 2 child comments replies: [–] squaresinger@lemmy.world 3 points 1 year ago (1 child) Moore's law hasn't been exponential for ~15 years now. permalink fedilink source parent hideshow 2 child comments replies: [–] Womble@lemmy.world 1 point 1 year ago If you are just talking transitor density I believe it still is, but even if not, my point was that it had exponential growth spanning over many decades. permalink fedilink source parent
[–] prole@lemmy.blahaj.zone 6 points 1 year ago and I can easily check it was completed correctly Can you always though? permalink fedilink source parent
[–] Womble@lemmy.world 1 point 1 year ago (1 child) That said, exponentials don’t exist in the real world, we’re just seeing the middle of a sigmoid curve, which will soon yield diminishing returns. Yes, but the tricky thing is we have no idea when the seemingly exponential growth will flip over into the plateuing phase. We could be there already or it could be another 30 years. For comparison Moores law is almost certainly a sigmoid too, but weve been seeing exponential growth for 50 years now. permalink fedilink source parent hideshow 2 child comments replies: [–] squaresinger@lemmy.world 3 points 1 year ago (1 child) Moore's law hasn't been exponential for ~15 years now. permalink fedilink source parent hideshow 2 child comments replies: [–] Womble@lemmy.world 1 point 1 year ago If you are just talking transitor density I believe it still is, but even if not, my point was that it had exponential growth spanning over many decades. permalink fedilink source parent
[–] squaresinger@lemmy.world 3 points 1 year ago (1 child) Moore's law hasn't been exponential for ~15 years now. permalink fedilink source parent hideshow 2 child comments replies: [–] Womble@lemmy.world 1 point 1 year ago If you are just talking transitor density I believe it still is, but even if not, my point was that it had exponential growth spanning over many decades. permalink fedilink source parent
[–] Womble@lemmy.world 1 point 1 year ago If you are just talking transitor density I believe it still is, but even if not, my point was that it had exponential growth spanning over many decades. permalink fedilink source parent
[–] spankmonkey@lemmy.world 1 point 1 year ago All that power used for a fucking coin flip. permalink fedilink source parent