Tag
A tweet highlights that frontier reinforcement learning is now an infrastructure problem, noting the use of the open-source slime library in Modal's RL stack and upstream contributions.