一作ACL 2026Findings· Findings of ACL
N- GRPO : Embedding-Level Neighbor Mixing for Enhanced Policy Optimization
Xukun Zhu · Hang Yu · Peng Di · Linchao Zhu
摘要
The success of Large Language Models in mathematical reasoning relies heavily on the generation of diverse and valid solution paths during the rollout phase. However, current rollout techniques face a fundamental trade-off: token-level sampling often yields redundant trajectories that differ only...