Multi-Domain Candidate Answer Routing for Privacy-Aware Assistants

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing virtual assistants struggle to provide answers to user questions in formats that are most useful to the user while maintaining privacy, as they often rely on a single information domain and lack integration with client-based context data.

Innovation Solution

A system that accesses multiple information domains, including server-based and client-based contexts, to provide candidate answers with confidence scores, allowing for selection of the most appropriate format based on user preferences and privacy-preserving data usage.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Adaptability or versatility

If a single information domain is used to provide answers, then the system complexity is reduced, but the answer format adaptability and user need satisfaction deteriorate

Engineering Contradiction:
Improveanswer format adaptabilityVSAvoidsystem complexity
Core Design Contradiction:
Adaptability or versatilityVSDevice complexity

Solution Approach 1:

The system segments the information domain into multiple specialized domains (e.g., local information domain for device-specific data, remote information domain for cloud-based data). Each domain handles specific types of information retrieval, allowing the system to adapt to different user needs while maintaining manageable complexity through modular architecture.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The virtual assistant application is designed with multi-functionality to operate across multiple information domains. It can universally handle different types of queries by routing them to appropriate domains (local or remote), providing diverse answer formats without requiring separate systems for each function.

Inventive Principle:
Principle #6Universality (Multi-functionality)

2Measurement precision

If client-based context data is integrated, then the answer relevance and personalization are improved, but the privacy protection requirements increase

Engineering Contradiction:
Improveanswer relevanceVSAvoidprivacy risk
Core Design Contradiction:
Measurement precisionVSObject-affected harmful factors

Solution Approach 1:

The system applies local quality by processing sensitive context data locally on the user's device rather than centralizing it. The local information domain handles personal context data (contacts, messages, calendar events) on-device, ensuring that personalized answers can be generated without transmitting private information to remote servers, thus maintaining both relevance and privacy.

Inventive Principle:
Principle #3Local quality

Solution Approach 2:

The virtual assistant application acts as an intermediary between the user's local data and the information retrieval system. It selectively processes context data locally and only transmits necessary non-sensitive information to remote domains, mediating between personalization needs and privacy protection requirements.

Inventive Principle:
Principle #24Intermediary (Mediator)

3Reliability

If multiple information domains are accessed, then the candidate answer quality and format diversity are improved, but the information retrieval time increases

Engineering Contradiction:
Improvecandidate answer qualityVSAvoidinformation retrieval time
Core Design Contradiction:
ReliabilityVSLoss of time

Solution Approach 1:

The system performs preliminary actions by pre-processing and caching frequently accessed information in the local information domain. Common context data and frequently used information are prepared in advance, allowing the virtual assistant to quickly retrieve answers for routine queries without accessing multiple domains, thus reducing retrieval time while maintaining answer quality.

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The system applies partial action by selectively accessing only the necessary information domains based on the specific query type. Rather than always querying all domains, it routes queries to the most appropriate domain (local or remote), performing just enough information retrieval to provide a satisfactory answer, thereby reducing unnecessary time loss.

Inventive Principle:
Principle #16Partial or excessive action

Data Source

PatentUS12423355B2Accessing multiple domains across multiple devices for candidate responses
Publication Date: 2025.09.23 APPLE INC
  • US12423355B2 patent drawing
  • US12423355B2 patent drawing
  • US12423355B2 patent drawing

AI summary

A device implementing a system for determining an answer to a question includes a processor configured to receive a question from a client device, and send the question to a first information domain comprising plural search subdomains, each of the plural search subdomains being configured to provide a respective candidate answer to the question in a format that differs from those provided by the other search subdomains of the plural search subdomains. The processor is further configured to receive, in response to sending the question, the candidate answer from each of the plural search subdomains, and provide at least one of the candidate answers to the client device, for providing an answer to the question based on the at least one of the candidate answers.