Hierarchical Lexical Graph for Enhanced Multi-Hop Retrieval
This addresses the challenge of complex, multi-hop question answering for users of RAG systems, representing a novel method for a known bottleneck rather than an incremental improvement.
The paper tackles the problem of multi-hop retrieval in Retrieval-Augmented Generation (RAG) systems, which struggle with piecing together answers across semantically distant documents, by introducing the Hierarchical Lexical Graph (HLG) and two complementary retrievers, resulting in an average relative improvement of 23.1% in retrieval recall and correctness over naive chunk-based RAG across five datasets.
Retrieval-Augmented Generation (RAG) grounds large language models in external evidence, yet it still falters when answers must be pieced together across semantically distant documents. We close this gap with the Hierarchical Lexical Graph (HLG), a three-tier index that (i) traces every atomic proposition to its source, (ii) clusters propositions into latent topics, and (iii) links entities and relations to expose cross-document paths. On top of HLG we build two complementary, plug-and-play retrievers: StatementGraphRAG, which performs fine-grained entity-aware beam search over propositions for high-precision factoid questions, and TopicGraphRAG, which selects coarse topics before expanding along entity links to supply broad yet relevant context for exploratory queries. Additionally, existing benchmarks lack the complexity required to rigorously evaluate multi-hop summarization systems, often focusing on single-document queries or limited datasets. To address this, we introduce a synthetic dataset generation pipeline that curates realistic, multi-document question-answer pairs, enabling robust evaluation of multi-hop retrieval systems. Extensive experiments across five datasets demonstrate that our methods outperform naive chunk-based RAG achieving an average relative improvement of 23.1% in retrieval recall and correctness. Open-source Python library is available at https://github.com/awslabs/graphrag-toolkit.