Physical Intelligence 团队对 Hi Robot 在实际任务中的表现进行了评估（如清理桌子、做三明治和购物），并与先前的方法进行了比较。结果表明，Hi Robot 在性能上优于 GPT-4o 和平面 VLA 策略。如下面的定量评估所示， Hi Robot 在指令跟随准确率上比 GPT-4o 高出 40%，表明它在对用户提示和实时观察的对齐方面有更强的能力。 此外，Hi ...

Robot learning and language acquisition represent an interdisciplinary endeavour combining robotics, artificial intelligence and cognitive science. This field studies how autonomous systems can ...

The Rho-alpha model incorporates sensor modalities such as tactile feedback and is trained with human guidance, says ...

Julian is a contributor and former staff writer at CNET. He's covered a range of topics, such as tech, crypto travel, sports and commerce. His past work has appeared at print and online publications, ...

Microsoft launched Rho-alpha in late January 2026, a robot model that uses vision, language, and touch sensors for two-armed tasks.

Microsoft has introduced a new artificial intelligence model aimed at pushing robots beyond controlled ...

Large language models like ChatGPT display conversational skills, but the problem is they don’t really understand the words they use. They are primarily systems that interact with data obtained from ...

Microsoft introduces Rho-alpha, an AI that understands human commands and controls two-armed robots with near-human precision.

Two researchers at UC Berkeley and ETH Zurich have harnessed the power of OpenAI’s GPT-4o large language model to teach cheap robot arms to clean up spills. It’s a clever demonstration of how AI ...

As generative AI tools like ChatGPT capture global attention, a new frontier is emerging—physical AI, or artificial intelligence that can interact with the real world. While large language models are ...

Google DeepMind demonstrated robots that can adapt to new environments without retraining, respond to natural language commands in real-time, and perform complex physical tasks with increased ...