作者位次 #2/15ACL 2026Poster· Natural Language Processing
OS -Symphony: A Holistic Framework for Robust and Generalist Computer-Using Agents
Bowen Yang · Kaiming Jin · Zhenyu Wu · Zhaoyang Liu · Qiushi Sun · Zehao Li · JingJing Xie · Zhoumianze Liu · Fangzhi Xu · Kanzhi Cheng · Yian Wang · Qingyun Li · Yu Qiao · Zun Wang · Zichen Ding
摘要
While Vision-Language Models (VLMs) have significantly advanced Computer-Using Agents (CUAs), current agentic frameworks struggle with robustness in novel domains and long-horizon workflows due to the absence of visual-aware tutorial retrieval and the lack of granular control over historical visu...