X · @teortaxesTex
· X / Twitter
Guys, look here please. RedNote's AI lab says they have a new RL regimen (TEMPO; beta version) that rewards world exploration allowing a 16B active Mo…
Guys, look here please.RedNote's AI lab says they have a new RL regimen (TEMPO; beta version) that rewards world exploration allowing a 16B active MoE to score >30% on ARC-AGI 3. «dots3-note Preview is an interim preview release. Reinforcement learning is not yet complete»dots studio: Introducing dots3-note preview — a