MDER-DR: Multi-Hop Question Answering with Entity-Centric Summaries

2026-03-11

AI Summary

This work addresses the limitations of existing knowledge graph–based retrieval-augmented generation (RAG) methods in multi-hop question answering, which suffer from performance degradation due to loss of contextual details when indexing triples. To overcome this, the authors propose the MDER-DR framework, which constructs entity-centric summaries through a Map-Disambiguate-Enrich-Reduce pipeline and integrates contextual triple descriptions to refine the retrieval index. Additionally, a Decompose-Resolve mechanism is introduced to iteratively decompose complex queries and perform stepwise reasoning without explicit graph traversal. Evaluated on both general and domain-specific benchmarks, the approach achieves up to a 66% improvement over standard RAG, demonstrating significantly enhanced robustness to sparse and intricate relational structures while maintaining consistent cross-lingual performance.

Technology Category

• Data Mining & Knowledge Management: Linked Open Data, Knowledge Graphs & KB Completion
• Natural Language Processing: Question Answering
• Knowledge Representation and Reasoning: Description Logics

Application Category

• Search and Retrieval-Augmented AI: Retrieval-Augmented Generation (RAG) and multi-modal RAG
• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs
• Graph Algorithms and Modeling for the Web: Querying, indexing, and retrieval in Web-related graphs

Abstract

Retrieval-Augmented Generation (RAG) over Knowledge Graphs (KGs) suffers from the fact that indexing approaches may lose important contextual nuance when text is reduced to triples, thereby degrading performance in downstream Question-Answering (QA) tasks, particularly for multi-hop QA, which requires composing answers from multiple entities, facts, or relations. We propose a domain-agnostic, KG-based QA framework that covers both the indexing and retrieval/inference phases. A new indexing approach called Map-Disambiguate-Enrich-Reduce (MDER) generates context-derived triple descriptions and subsequently integrates them with entity-level summaries, thus avoiding the need for explicit traversal of edges in the graph during the QA retrieval phase. Complementing this, we introduce Decompose-Resolve (DR), a retrieval mechanism that decomposes user queries into resolvable triples and grounds them in the KG via iterative reasoning. Together, MDER and DR form an LLM-driven QA pipeline that is robust to sparse, incomplete, and complex relational data. Experiments show that on standard and domain specific benchmarks, MDER-DR achieves substantial improvements over standard RAG baselines (up to 66%), while maintaining cross-lingual robustness. Our code is available at https://github.com/DataSciencePolimi/MDER-DR_RAG.

Problem

• Retrieval-Augmented Generation
• Knowledge Graphs
• Multi-Hop Question Answering
• Contextual Nuance
• Entity-Centric Summaries

Innovation

• Retrieval-Augmented Generation
• Knowledge Graphs
• Multi-Hop Question Answering
• Entity-Centric Summarization
• Iterative Reasoning