ACE-GraphRAG: Agentic Context Engineering for Hierarchical GraphRAG

2026-08-02

AI Summary

This work addresses the representation–reasoning gap in Hierarchical GraphRAG, where fixed context construction fails to align with dynamic query and task requirements. To bridge this gap, the authors propose introducing a contextual strategy layer at inference time, treating context assembly as a query- and task-aware dynamic reasoning process. Two strategies—Full-ACE and Adaptive-ACE—are designed, integrating parallel divergent retrieval (via depth-oriented factual and breadth-oriented semantic branches), incremental context fusion, and task-adaptive selection mechanisms to enable dynamic multi-granularity knowledge integration. Experimental results demonstrate that the proposed approach significantly outperforms existing RAG and GraphRAG baselines on HotpotQA, 2WikiMultiHopQA, and four UltraDomain subsets, with Adaptive-ACE achieving superior performance and higher human preference in multi-hop question answering and domain summarization tasks.

Abstract

Hierarchical Graph Retrieval-Augmented Generation (GraphRAG) organizes corpus knowledge at multiple levels of granularity, yet fixed context construction may fail to translate these multi-resolution representations into a context suited to the current query. We identify this mismatch as the representation--inference gap. We propose Agentic Context Engineering for Hierarchical GraphRAG (ACE-GraphRAG), an inference-time context policy layer that supplements and adapts the initial context for generation. ACE-GraphRAG formulates context construction as a policy over gap-aware refinement, retrieval branches, and task-conditioned adaptation. Parallel Differential Retrieval acquires supplementary evidence from depth-oriented factual and breadth-oriented semantic branches. These evidence increments are consolidated with the initial context while preserving provenance and abstraction levels. Full-ACE applies the full policy uniformly within each task family, whereas Adaptive-ACE selects task- and topology-specific policies for individual queries. We evaluate ACE-GraphRAG on HotpotQA, 2WikiMultiHopQA, and four UltraDomain subsets across multi-hop QA and query-focused summarization. Full-ACE outperforms the evaluated RAG and GraphRAG baselines across both task families, while Adaptive-ACE further improves multi-hop QA and is preferred over Full-ACE on all four UltraDomain subsets. Ablation and topology analyses support treating context construction as a query- and task-dependent inference policy rather than a fixed procedure.

Problem

• Hierarchical GraphRAG
• representation-inference gap
• context construction
• multi-resolution knowledge
• query-adaptive inference

Innovation

• Agentic Context Engineering
• Hierarchical GraphRAG
• Parallel Differential Retrieval
• Representation–Inference Gap
• Task-Conditioned Adaptation