RAG / 检索增强生成

PRISMA: Reinforcement Learning Guided Two-Stage Policy Optimization in Multi-Agent Architecture for Open-Domain Multi-Hop Question Answering

PRISMA：基于强化学习的多智能体架构中两阶段策略优化

Yu Liu, Wenxiao Zhang, Cong Cao, Wenxuan Lu, Fangfang Yuan, Diandian Guo, Kun Peng, Qiang Sun, Kaiyan Zhang, Yanbing Liu, Jin B. Hong, Bowen Zhou, Zhiyuan Ma

专题命中 知识库问答 ：retrieval-augmented generation(abstract)；RAG(abstract)；分类 cs.AI

AI总结 PRISMA通过两阶段组相对策略优化，结合推理引导协作，解决多跳问答中的检索崩溃和学习不稳定问题，实现高性能和稳定部署。