一作ACL 2026Findings· Findings of ACL
BAPO : Boundary-Aware Policy Optimization for Reliable Agentic Search
Shiyu Liu · Yongjing Yin · Jianhao Yan · Yunbo Tang · Qinggang Zhang · Bei Li · Xin Chen · Jingang Wang · Xunliang Cai · Jinsong Su
摘要
RL-based agentic search enables LLMs to solve complex questions via dynamic planning and external search. While this approach significantly enhances accuracy with agent policies optimized via large-scale reinforcement learning, we identify a critical gap in reliability: these agents fail to recog...