Ai2’s new Olmo 3.1 extends reinforcement learning training for stronger reasoning benchmarks

Ai2’s new Olmo 3.1 extends reinforcement learning training for stronger reasoning benchmarks