Digital Assistant Query Routing With Context-Aware Rewriting
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing digital assistants face inefficiencies in routing queries to appropriate components and processing user requests, leading to increased latency and power consumption, especially in battery-powered devices.
Innovation Solution
Implementing a system that determines the type of query and routes it to the most efficient digital assistant component, leveraging large language models and multiple handling agents to process queries efficiently, and adjusting queries based on contextual data.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If a single digital assistant component processes all queries, then the system is simple to manage, but processing latency increases and power consumption rises
Solution Approach 1:
The patent divides the digital assistant system into multiple specialized components: a query routing component that classifies incoming queries, and multiple handling agents (first handling agent, second handling agent, third handling agent) that process different query types. This segmentation allows each component to be optimized for its specific function, improving overall processing efficiency while maintaining manageable complexity through modular design.
2Measurement precision
If all queries are processed with full contextual analysis, then accuracy improves, but power consumption increases significantly
Solution Approach 1:
The query routing component performs preliminary classification of queries before they reach the handling agents. By analyzing the query type upfront and routing it to the appropriate specialized agent, the system avoids performing full contextual analysis on every query. This preliminary action ensures accurate routing while reducing overall power consumption by limiting deep analysis to only when necessary.
Solution Approach 2:
Different handling agents are assigned to process different query types (e.g., factual queries, conversational queries, task-oriented queries). Each agent applies contextual analysis at the appropriate level for its specific query type, rather than all agents performing identical comprehensive analysis. This local quality approach ensures sufficient accuracy for each query type while optimizing power consumption by avoiding unnecessary deep analysis.
3Measurement precision
If the system rewrites all queries based on contextual data, then response accuracy improves, but processing time increases
Solution Approach 1:
The system applies query rewriting based on contextual data selectively rather than universally. The routing component determines which queries benefit from rewriting and directs only those to the appropriate handling agents. This partial action approach maintains high response accuracy for queries that need rewriting while minimizing processing latency by avoiding unnecessary rewriting operations on queries that can be handled directly.
Data Source
AI summary
Systems and processes for operating an intelligent automated assistant are provided. An example method includes, at a computer system that is configured to communicate with a display generation component and an input device: detecting an audio input including a query; in response to detecting the audio input including the query: retrieving contextual data related to the query; in accordance with a determination that the query includes a request of a first type: converting the query to a rewritten query based on the contextual data related to the query; and providing the rewritten query to a first digital assistant component; and in accordance with a determination that the query includes a request of a second type different from the request of the first type, providing the query and the contextual data related to the query to a second digital assistant component different from the first digital assistant component.