Transparent and Controllable Human-Ai Interaction Via Chaining of Machine-Learned Language Models

    公开(公告)号:US20250036376A1

    公开(公告)日:2025-01-30

    申请号:US18915020

    申请日:2024-10-14

    Applicant: Google LLC

    Abstract: The present disclosure provides to transparent and controllable human-AI interaction via chaining of machine-learned language models. In particular, although existing language models (e.g., so-called “large language models” (LLMs)) have demonstrated impressive potential on simple tasks, their breadth of scope, lack of transparency, and insufficient controllability can make them less effective when assisting humans on more complex tasks. In response, the present disclosure introduces the concept of chaining instantiations of machine-learned language models (e.g., LLMs) together, where the output of one instantiation becomes the input for the next, and so on, thus aggregating the gains per step.

    FRONT-END USER INTERFACE DESIGN TOOL AND HUMAN READABLE CODE GENERATOR

    公开(公告)号:US20250013440A1

    公开(公告)日:2025-01-09

    申请号:US18218048

    申请日:2023-07-04

    Abstract: Systems and method for automatically generating front-end code for a user interface (UI) design created in a graphical UI editor by receiving a design file key and an access token associated with the design file; using the design file key and access token to retrieve one or more outermost frames within a project; obtaining thumbnails of the one or more frames; enabling user selection of the one or more frames for import and retrieving the selected frames along with their components; retrieving image files used within the one or more frames; mapping the design elements to code templates for one or more front-end frameworks; generating front-end code based on the mapped design elements and code templates; and outputting the generated front-end code in a format suitable for use in a web or mobile application.

    Development of multiple classes of user interfaces for a software application

    公开(公告)号:US12182544B2

    公开(公告)日:2024-12-31

    申请号:US18199307

    申请日:2023-05-18

    Inventor: Allen Bauer

    Abstract: In a method for displaying a user interface at a device, a software application is received at the device, wherein the software application includes multiple classes of user interfaces pertaining to different versions of the software application executing on different device platforms having at least a same form factor or a same operating system, wherein the software application includes a master view and a plurality of subordinate views, wherein each of the subordinate views pertain to one of the multiple classes of user interfaces. The software application is executed at the device. An appropriate user interface of the multiple classes of user interfaces is determined for device based on at least a form factor of the device. The appropriate user interface is displayed at the device by displaying the master view and the subordinate view for the appropriate user interface.

    PROTOTYPING APPLICATIONS OF SPATIALLY AWARE SMART OBJECTS USING AUGMENTED REALITY

    公开(公告)号:US20240419407A1

    公开(公告)日:2024-12-19

    申请号:US18335559

    申请日:2023-06-15

    Abstract: One or more devices, systems, methods and/or non-transitory, machine-readable mediums are described herein for prototyping applications of spatially aware smart objects using augmented reality (AR). In an embodiment, a system can comprise a memory that stores computer-executable components, and a processor that executes the computer-executable components stored in the memory. The computer-executable components comprise a spatial detection component that determines and tracks spatial positions and orientations of one or more objects in association with moving the one or more objects within a real-world environment, and an interface component comprising a visual programming user interface that facilitates prototyping spatial events and corresponding effects associated with the moving of one or more objects based on the relative spatial positions.

    Robotic process automation
    10.
    发明授权

    公开(公告)号:US12153904B2

    公开(公告)日:2024-11-26

    申请号:US17350084

    申请日:2021-06-17

    Abstract: A method, system, and computer program product for automated identification and generation of user interface interactions for robotic process automation development are provided. The method identifies an interface environment including a set of user interface elements. The method identifies a set of element attributes associated with the set of user interface elements based on a user interface model. A set of automated interactions are generated based on the interface environment, the set of user interface elements, the set of element attributes, and the user interface model. An input interface is generated based on the set of automated interactions. The method generates program code for a thematic interface environment based on user input received through the input interface, the interface environment, the set of user interface elements, the set of element attributes, and the user interface model.

Patent Agency Ranking