The ReQAP System for Question Answering over Personal Information

2025-08-09

AI Summary

This work addresses complex querying over heterogeneous, multi-source personal data (e.g., calendars, emails, social posts, fitness logs) directly on end-user devices. Methodologically, it introduces a traceable, recursive query execution framework that decomposes natural language questions hierarchically into sub-questions and incrementally constructs an executable operator tree; a lightweight language model—finetuned end-to-end—jointly performs semantic parsing and cross-source data operations (filtering, joining, aggregation). Its key contributions are: (i) the first integration of recursive decomposition with explicit operator-tree construction for personal data querying, enabling full end-to-end interpretability; and (ii) unified handling of both structured and unstructured multimodal data. Experiments demonstrate significant improvements in answer accuracy for complex queries and provide complete, step-by-step reasoning traceability—enhancing user trust and controllability.

Technology Category

Data Mining & Knowledge Management: Intelligent Query ProcessingNatural Language Processing: Question AnsweringSearch and Optimization: Distributed Search

Application Category

Search and Retrieval-Augmented AI: Personalized, context-aware and across-device searchUser Modeling, Personalization and Recommendation: Fairness-aware retrieval and rankingSemantics and Knowledge: Data modeling to support human-machine intelligence, including LLMs agents, intelligent system behavior, explanations, and user-friendly interactions

Abstract

Personal information is abundant on users' devices, from structured data in calendar, shopping records or fitness tools, to unstructured contents in mail and social media posts. This works presents the ReQAP system that supports users with answers for complex questions that involve filters, joins and aggregation over heterogeneous sources. The unique trait of ReQAP is that it recursively decomposes questions and incrementally builds an operator tree for execution. Both the question interpretation and the individual operators make smart use of light-weight language models, with judicious fine-tuning. The demo showcases the rich functionality for advanced user questions, and also offers detailed tracking of how the answers are computed by the operators in the execution tree. Being able to trace answers back to the underlying sources is vital for human comprehensibility and user trust in the system.

Problem

• Answering complex questions over personal data
• Decomposing queries across heterogeneous sources
• Providing traceable answers for user trust

Innovation

• Recursively decomposes complex questions for processing
• Incrementally builds operator tree for execution
• Uses light-weight language models with fine-tuning