Hacker News new | ask | show | jobs
by ethbr1 132 days ago
Isn't that just RL with extra power-intensive steps? (An entire model chugging away in the goal function)
1 comments
hrn_frs 132 days ago
That's correct, but if successful you'd essentially have updated the LLM's knowledge and capabilities "on the fly".
link
ethbr1 131 days ago
Maybe we could run off-peak load of that nature, when power is cheaper. Call it dreaming. ;)
link