Outcome-reward reinforcement learning such as GRPO gives every token of a search-agent rollout the same advantage, although the rollout exercises two capabil…
机构：腾讯
来源：arXiv 2610.23148 | AI4Papers 论文推荐平台