作者位次 #2/11ACL 2026Poster· Natural Language Processing
Experience-driven Multi-turn Reinforcement Learning for GUI Agents
Zhengxi Lu · Jiabo Ye · Fei Tang · Yongliang Shen · Haiyang Xu · Ziwei Zheng · Weiming Lu · Ming Yan · Fei Huang · Jun Xiao · Yueting Zhuang
摘要
GUI agents have demonstrated remarkable progress in automating complex user interface interactions. However, training such agents for long-horizon tasks remains challenging. Single-turn reinforcement learning conditions on expert histories during training but self-generated histories during deplo...