作者位次 #2/5ACL 2026Findings· Findings of ACL
Hybrid Self-evolving Structured Memory for Computer-Use Agents
Sibo Zhu · Wenyi WU · Kun Zhou · Stephen Wang · Biwei Huang
摘要
The remarkable progress of vision–language models (VLMs) has enabled computer-use agents to interact with computers in a human-like manner. Yet real-world computer-use tasks remain difficult due to long-horizon workflows, diverse interfaces, and frequent intermediate errors. Prior work equips age...