6 citations · 7 across the 4 of their papers we have counts for
1 paper · 1 filter
Sihwan Park, Sung-Yub Kim, Doohyuk Jang +1
Asynchronous rollout improves the efficiency of reinforcement learning with verifiable rewards (RLVR) by reducing under-utilization during long-tail generation. Sustained concurren…