作者位次 #7/10NeurIPS 2026Poster· Reinforcement Learning / Deep RL
Learn from the Gap: Differential-Aware Advantage Pruning with Adaptive Rollout Sampling for GRPO
Jiahua Yang · Chuangchuang Wang · Zhiwei Yang · Xianpeng Zhang · Dongyu Chen · Xing Chen · Tianhuang Su · Haonan Lu · Quanlong Guan · Kai Tang