Efficiently Extendable In-Interpreter Natural Language Agent

    公开(公告)号:US20240346246A1

    公开(公告)日:2024-10-17

    申请号:US18300930

    申请日:2023-04-14

    申请人: ServiceNow, Inc.

    IPC分类号: G06F40/279

    CPC分类号: G06F40/279

    摘要: A trained natural language model is provided that uses an input session history to generate outputs to an interpreter. Outputs to the interpreter, and inputs responsively received therefrom, are added to the history to generate additional model outputs as the history is updated. The model is trained to engage in goal-oriented dialog with the interpreter and with the user (optionally through interpreter function calls) to identify the user's goals, to learn information about modules, functions, and methods available in the interpreter that are relevant to the user's goals, and to execute function calls and/or commands, based on the learned information, that accomplish the user's goals. The use of a history that may be completely blank at the beginning of the session reduces the computational requirements of running the model, as well as allowing the model to ‘update’ itself as the available modules are update, added, or removed.

    Server-side workflow improvement based on client-side data mining

    公开(公告)号:US12095842B2

    公开(公告)日:2024-09-17

    申请号:US18375757

    申请日:2023-10-02

    申请人: ServiceNow, Inc.

    IPC分类号: H04L67/01 H04L67/50

    CPC分类号: H04L67/01 H04L67/535

    摘要: An embodiment may involve a server-side log collected by a server device, where the server-side log includes a set of entries indicating a unique identifier, and wherein the unique identifier is assigned to a work item of a server-based application executed by the server device. The embodiment may further involve one or more processors configured to: receive, from a client device disposed upon a network, a client-side log, wherein the client-side log includes operational data related to usage of a client-based application executed by the client device; identify, from the operational data, the client-based application and one or more activities performed by the client-based application; determine that the one or more activities are related to the unique identifier; based on the one or more activities, determine an action that can be taken to improve efficacy of the server-based application; and write, to the persistent storage, a representation of the action.

    SYSTEM AND METHOD FOR DELIVERING APPLICATION METADATA

    公开(公告)号:US20240303293A1

    公开(公告)日:2024-09-12

    申请号:US18666023

    申请日:2024-05-16

    申请人: ServiceNow, Inc.

    IPC分类号: G06F16/957 H04L67/02

    CPC分类号: G06F16/9574 H04L67/02

    摘要: A server receives a first hypertext transfer protocol (HTTP) from a client device requesting for content associated with a webpage. The server retrieves a plurality of cache keys associated with respective sets of application metadata identified in the first HTTP request from an application metadata database and transmits the cache keys to the client device. The server receives a second HTTP request from the client device identifying one or more cache keys that are not stored in a local HTTP cache of the client device. The server device retrieves the sets of application metadata corresponding to the missing cache keys from the application metadata database and transmits the application metadata to the client device.

    Orchestration for Robotic Process Automation
    10.
    发明公开

    公开(公告)号:US20240264815A1

    公开(公告)日:2024-08-08

    申请号:US18638296

    申请日:2024-04-17

    申请人: ServiceNow, Inc.

    IPC分类号: G06F8/60 G06F9/451

    CPC分类号: G06F8/60 G06F9/451

    摘要: An example embodiment may involve persistent storage containing definitions of (i) assignments of bots to endpoints, (ii) software packages for execution by the bots, (iii) schedules for the bots to execute the software packages, and (iv) processes that associate the bots with the software packages and the schedules. This embodiment may also involve processors configured to: receive, from a computing device, a request for bot deployment, wherein the computing device includes a bot runtime; identify, in the processes, a bot assigned to an endpoint that is associated with the computing device, wherein the bot is associated with a software package and a schedule; and transmit, to the computing device, data including a representation of the bot, a copy of the software package, and a copy of the schedule, wherein reception of the data causes the bot to execute, using the bot runtime, the software package in accordance with the schedule.