aiqbench analysis
This research introduces HCG-RAG, a method that constructs compact causal graphs using fixed schemas to significantly reduce the computational cost of GraphRAG systems. By filtering retrieval based on hierarchical structure rather than exhaustive entity extraction, it achieves superior efficiency and accuracy over flat embedding-only baselines across medical benchmarks. The study demonstrates that imposing higher-level organizational constraints yields better performance than simply scaling graph size with corpus length.
Structure Over Scale: Schema-Constrained Causal Graphs for RAG
arXiv:2607.22592v1 Announce Type: new Abstract: Graph-based retrieval-augmented generation (GraphRAG) grounds answers in structured knowledge, but current systems extract entities and relationships exhaustively, producing graphs whose size and construction cost scale with corpus length rather than with the reasoning a query requires. We introduce HCG-RAG (Hierarchical Causal Graph RAG), which replaces open-ended extraction with schema-constrained causal graphs: an automated pipeline distills a corpus into a fixed, typed vocabulary of causal variables and materializes a compact two-tier graph over it. Our schema-constrained graphs match entity-relation baselines on answer quality at a fraction of the cost: 3-20x fewer nodes, 8x-135x fewer build-time LLM calls than the most LLM-intensive baseline (MS-GraphRAG), and graphs compact enough for a domain expert to audit, correct, and extend. On medical and clinical benchmarks, including a neurologist-validated epilepsy dataset, HCG-RAG matches or exceeds the best entity-relation systems. An ablation isolates the causal graph as a structured retrieval filter, contributing +6 percentage points (pp) over embedding-only retrieval. Across all domains with discoverable hierarchical causal structure, only methods imposing higher-level organization outperform flat entity-relation retrieval, indicating that what is placed in the graph matters more than how many nodes it contains.
Abstract:Graph-based retrieval-augmented generation (GraphRAG) grounds answers in structured knowledge, but current systems extract entities and relationships exhaustively, producing graphs whose size and construction cost scale with corpus length rather than with the reasoning a query requires. We introduce HCG-RAG (Hierarchical Causal Graph RAG), which replaces open-ended extraction with schema-constrained causal graphs: an automated pipeline distills a corpus into a fixed, typed vocabulary of causal variables and materializes a compact two-tier graph over it. Our schema-constrained graphs match entity-relation baselines on answer quality at a fraction of the cost: 3-20x fewer nodes, 8x-135x fewer build-time LLM calls than the most LLM-intensive baseline (MS-GraphRAG), and graphs compact enough for a domain expert to audit, correct, and extend. On medical and clinical benchmarks, including a neurologist-validated epilepsy dataset, HCG-RAG matches or exceeds the best entity-relation systems. An ablation isolates the causal graph as a structured retrieval filter, contributing +6 percentage points (pp) over embedding-only retrieval. Across all domains with discoverable hierarchical causal structure, only methods imposing higher-level organization outperform flat entity-relation retrieval, indicating that what is placed in the graph matters more than how many nodes it contains.
Artificial Intelligence (cs.AI); Computation and Language (cs.CL); Information Retrieval (cs.IR)
arXiv:2607.22592 [cs.AI]
(or arXiv:2607.22592v1 [cs.AI] for this version)
https://doi.org/10.48550/arXiv.2607.22592 Focus to learn more arXiv-issued DOI via DataCite (pending registration)
arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.
Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.
Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.
Full text under the publisher's open licence · 286 words
Related coverage
Presentation: Ontology‐Driven Observability: Building the E2E Knowledge Graph at Netflix Scale
InfoQ · Oct 9, 2026
The Entangled Dimensions of Decision Theory
LessWrong · Jul 30, 2026
Externalized CPDAG Summaries Improve LLM Causal Deduction
arXiv · Sep 28, 2026
Tracing the Cascade: A Topology-Aware Evaluation Framework for Scientific Agent Hallucinations
arXiv · Aug 5, 2026