ChatGLM has announced the launch of AutoGLM, a new technology aimed at enabling autonomous foundation agents for graphical user interfaces (GUIs), particularly focusing on web browsers and Android devices. The beta testing for AutoGLM began on October 25, 2024, showcasing its capabilities in phone use. The innovation is expected to revolutionize AI interactions with mobile technology, marking a significant advancement in the ChatGLM family. The project emphasizes the importance of designing an intermediate interface that separates planning and grounding behaviors in GUI agents, achieving state-of-the-art results in its applications.
🚀 Introducing AutoGLM! A new milestone in the ChatGLM family, AutoGLM is here to enable foundation agents for autonomous control of GUIs, from web browsers to smartphones. https://t.co/kJ6sOJeGIN
🏷️:OS-ATLAS: A Foundation Action Model for Generalist GUI Agents 🔗:https://t.co/VsEj2C6xuO https://t.co/hcogsg9f3Z
AutoGLM: Autonomous Foundation Agents for GUIs Focuses on Web Browser and Android as the representative GUI scenarios. Found it essential to design an intermediate interface that disentangles planning and grounding behaviors in foundation GUI agents. Achieves SOTA results in… https://t.co/xuaklITdDT