Xinyu AI Search: Enhanced Relevance and Comprehensive Results with Rich Answer Presentations

2025-05-28

AI Summary

Traditional search engines struggle to integrate fragmented information for complex queries, while existing generative AI search systems suffer from limitations in relevance, comprehensiveness, and result presentation. To address these challenges, we propose a Query Decomposition Graph (QDG)-driven collaborative retrieval-generation framework: complex queries are modeled as executable decomposition graphs enabling stepwise retrieval, multi-source heterogeneous fusion, dynamic filtering, and adaptive re-ranking; a fine-grained inline citation mechanism ensures traceable and trustworthy provenance; and we introduce novel timeline-based visualization and text-image co-rendering paradigms. In evaluations on real-world complex queries, our system outperforms eight baseline models across all metrics. Human evaluation confirms significant improvements in relevance, comprehensiveness, and insight generation. Ablation studies validate the critical contribution of each component.

Technology Category

• Search and Optimization: Distributed Search
• Data Mining & Knowledge Management: Conversational Systems for Recommendation & Retrieval
• Computer Vision: Image and Video Retrieval

Application Category

• Search and Retrieval-Augmented AI: Retrieval-Augmented Generation (RAG) and multi-modal RAG
• Graph Algorithms and Modeling for the Web: Querying, indexing, and retrieval in Web-related graphs
• User Modeling, Personalization and Recommendation: Fairness-aware retrieval and ranking

Abstract

Traditional search engines struggle to synthesize fragmented information for complex queries, while generative AI search engines face challenges in relevance, comprehensiveness, and presentation. To address these limitations, we introduce Xinyu AI Search, a novel system that incorporates a query-decomposition graph to dynamically break down complex queries into sub-queries, enabling stepwise retrieval and generation. Our retrieval pipeline enhances diversity through multi-source aggregation and query expansion, while filtering and re-ranking strategies optimize passage relevance. Additionally, Xinyu AI Search introduces a novel approach for fine-grained, precise built-in citation and innovates in result presentation by integrating timeline visualization and textual-visual choreography. Evaluated on recent real-world queries, Xinyu AI Search outperforms eight existing technologies in human assessments, excelling in relevance, comprehensiveness, and insightfulness. Ablation studies validate the necessity of its key sub-modules. Our work presents the first comprehensive framework for generative AI search engines, bridging retrieval, generation, and user-centric presentation.

Problem

• Improves relevance and comprehensiveness in AI search results
• Dynamically decomposes complex queries for stepwise retrieval
• Innovates result presentation with citations and visualizations

Innovation

• Query-decomposition graph for stepwise retrieval
• Multi-source aggregation enhances diversity
• Timeline visualization improves result presentation