阿里研发的Mobile-Agent框架实现了智能手机的跨应用操作。该框架基于多模态大模型,通过视觉分析实现应用的智能控制,无需人工编写脚本。用户可以用语音或文字指示Mobile-Agent在多个应用之间完成指定任务。
例如,用户可以让Mobile-Agent在浏览器中搜索比赛结果,然后在备忘录中撰写相关报道。这打破了应用间的界限,使智能手机助手真正具备跨应用的协同能力。
Mobile-Agent的推出是智能手机向智能体发展的重要一步。它不仅增强了手机的智能化,也为人机交互带来更多可能性。
Title: Alibaba Launches Mobile-Agent for Smartphone Cross-app Operation
Keywords: Smartphone, Cross-app, Intelligent Agent
News content:
Alibaba has developed the Mobile-Agent framework that enables cross-app operation on smartphones. Based on multimodal foundation models, the framework realizes intelligent control of apps through visual perception without manual scripting. Users can instruct Mobile-Agent to complete designated tasks across multiple apps via voice or text.
For example, users can ask Mobile-Agent to search for match results in the browser and then draft related reports in the memo app. This breaks the barriers between apps and truly empowers the smartphone assistant with cross-app collaborative capabilities.
The launch of Mobile-Agent is an important step towards transforming smartphones into intelligent agents. It not only enhances the intelligence of smartphones but also brings more possibilities to human-machine interaction.
【来源】https://www.qbitai.com/2024/02/118426.html
Views: 2