Task processing method and device of intelligent agent, electronic equipment, computer readable storage medium and computer program product

By displaying the agent's current and historical decision summaries in the task processing interface and providing rollback trajectories in abnormal states, the problem of users having difficulty distinguishing steps and recovering complexity in agent task processing is solved, achieving fast and accurate task recovery.

CN122242555APending Publication Date: 2026-06-19TENCENT TECH (BEIJING) CO LTD
View PDF 0 Cites 0 Cited by

Patent Information

Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
TENCENT TECH (BEIJING) CO LTD
Filing Date
2026-03-16
Publication Date
2026-06-19

AI Technical Summary

Technical Problem

In the process of intelligent agent task processing, existing technologies make it difficult for users to quickly distinguish between completed key stages and the current steps in progress, and lack an intuitive backtracking mechanism, which leads to complex recovery from abnormal states, affecting the success rate of task processing and user experience.

Method used

The task processing interface displays the agent's current execution path and multiple decision summaries, including historical and current decision summaries, and provides diagnostic information and rollback trajectories in abnormal states, allowing users to directly roll back to the target node through operations.

Benefits of technology

With clear decision summaries and rollback mechanisms, users can quickly understand task progress and perform precise recovery in abnormal situations, improving the robustness of task processing and the efficiency of anomaly recovery.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN122242555A_ABST
    Figure CN122242555A_ABST
Patent Text Reader

Abstract

This application provides a task processing method, apparatus, electronic device, computer-readable storage medium, and computer program product for an intelligent agent. The method includes: displaying the intelligent agent's thought process for the current execution path of a target task in a task processing interface, and displaying multiple decision summaries; if an abnormal state occurs during the execution of the current execution path, displaying diagnostic information for the abnormal state, wherein the diagnostic information includes the target decision summary and displays a backtracking trajectory from the abnormal state to the target decision summary; in response to an application operation on the backtracking trajectory, backtracking from the current execution node of the current execution path to the target node, wherein the current execution node is the latest node in the current execution path, and the target node is the completed node corresponding to the target decision summary and whose execution order is the last. This application improves the efficiency of the intelligent agent in handling abnormal states during task execution.
Need to check novelty before this filing date? Find Prior Art