Iterative task execution method and system based on dynamic feedback and causal fault tolerance

By constructing a task semantic graph and a causal fault tolerance mechanism, the problems of decomposition deviation and error propagation of complex tasks in enterprise-level intelligent data analysis platforms are solved, achieving adaptive task decomposition and closed-loop optimization, thereby improving the task execution success rate and system robustness.

CN121301846BActive Publication Date: 2026-06-09BEIJING SILICONFLOW TECHNOLOGY CO LTD
View PDF 2 Cites 0 Cited by

Patent Information

Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
BEIJING SILICONFLOW TECHNOLOGY CO LTD
Filing Date
2025-10-15
Publication Date
2026-06-09

AI Technical Summary

Technical Problem

Existing enterprise-level intelligent data analysis platforms suffer from problems such as the ambiguity of natural language expressions in the initial task decomposition, the difficulty in unifying semantic modeling of multimodal information, and the structural propagation of errors in the task chain when handling complex, multi-stage data analysis tasks. These problems result in low task execution success rates and poor system robustness.

Method used

An iterative data analysis task execution method based on dynamic feedback and causal fault tolerance is adopted. A task semantic graph is constructed by multimodal task information, and a graph neural network and attention mechanism are used for embedding representation. Intermediate results are monitored in real time and dynamic re-decomposition is triggered. An error propagation causal graph is constructed by combining historical execution trajectories, and a fault tolerance strategy is preset to optimize the task decomposition structure.

Benefits of technology

It significantly improves the success rate and robustness of complex data analysis tasks, achieves adaptive task decomposition and closed-loop optimization, avoids decomposition bias caused by information fragmentation, predicts high-risk paths and automatically embeds defensive operations to block the propagation of errors.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN121301846B_ABST
    Figure CN121301846B_ABST
Patent Text Reader

Abstract

This invention discloses an iterative task execution method and system based on dynamic feedback and causal fault tolerance, belonging to the field of artificial intelligence and data analysis technology. By integrating multimodal perception, dynamic feedback, and causal reasoning, it significantly improves the success rate and robustness of complex data analysis tasks, avoids decomposition bias caused by information fragmentation, quantifies task complexity through graph neural networks to achieve adaptive task decomposition, and combines a dynamic feedback mechanism to correct the task structure in real time during execution, supporting the insertion, merging, and order adjustment of subtasks. A causal fault tolerance strategy based on historical failure trajectories is introduced, which can predict high-risk paths and automatically embed defensive operations to block error propagation. The entire execution trajectory is stored in a memory bank for continuous optimization of the model and strategy, forming a closed-loop learning capability.
Need to check novelty before this filing date? Find Prior Art