Iris, a chatbot developed at Stanford University in California, can handle complex conversations, and could pave the way for digital assistants that understand how we really speak. Iris imitates this human technique by turning language commands into blocks of text that can be flexibly combined. This allows user requests to be tagged with instructions on how it can be stitched to further commands. This narrows the range of other commands that the tool can act on in the context of the conversation.
© 2001-2024 Fundación Dialnet · Todos los derechos reservados