1 paper · 1 filter
Shijie Zhang, Zheng Xiao, Shiyu Liu +7
Online reinforcement learning with verifiable rewards (RLVR) has become an effective paradigm for improving the reasoning abilities of large language models, but most methods still…