Search papers, labs, and topics across Lattice.
This paper introduces FlowRAG, a novel framework that enhances graph-based retrieval-augmented generation by constructing a quad-level heterogeneous graph that integrates passages, summaries, sentences, and entities. By employing a dual-granularity activation module and a frequency-aware weighted flow mechanism, FlowRAG improves semantic recall and enables more reliable multi-hop reasoning, addressing the limitations of existing entity-based graph methods. Experimental results demonstrate that FlowRAG achieves state-of-the-art performance on complex reasoning benchmarks, showcasing its effectiveness in knowledge-intensive tasks.
FlowRAG transforms how we approach multi-hop reasoning by leveraging a quad-level graph structure that enhances both semantic recall and explicit reasoning paths.
Graph-based retrieval-augmented generation (GraphRAG) is effective for knowledge-intensive and multi-hop query tasks; however, many existing methods primarily seed entity-based graphs and rely on implicit semantic relevance propagation. This often (i) under-retrieves when user queries are abstract and semantically sparse at the entity level, and (ii) suffers from brittle multi-hop reasoning, where noisy activations can derail entity-to-entity transitions and corrupt the inferred relation chain, yielding unreliable conclusions. To this end, we propose \texttt{FlowRAG}, a semantic-aware retrieval framework that improves both semantic recall and explicit reasoning. Specifically, \texttt{FlowRAG} constructs a quad-level heterogeneous graph over passages, summaries, sentences, and entities, where summary nodes serve as a coarse semantic hub. At retrieval time, a dual-granularity activation module combines summary--query alignment with sentence-level matching to activate relevant entities under paraphrase and abstraction robustly. We then introduce a frequency-aware weighted flow module that routes relevance through entity--passage links weighted by within-passage term frequency, pruning noisy connections and extracting high-confidence reasoning paths as an explicit logic skeleton for generation. Extensive experiments show that \texttt{FlowRAG} obtains state-of-the-art performance on complex reasoning benchmarks.
