Paper deep dive
ACE-GraphRAG: Agentic Context Engineering for Hierarchical GraphRAG
Yongfeng Huang, Yuren Lai, Ruiying Chen, Haoyu Huang, Mingming Zhao, James Cheng
Intelligence
Status: succeeded | Model: Gemma-4-26B-A4B | Prompt: intel-v1 | Confidence: 93%
Last extracted: 8/5/2026, 4:01:08 AM
Summary
The paper proposes ACE-GraphRAG, an inference-time context policy layer for Hierarchical GraphRAG that addresses the 'representation-inference gap' by adapting context construction to specific queries. It introduces Parallel Differential Retrieval for supplementary evidence and compares Full-ACE (uniform policy) with Adaptive-ACE (task-specific policy), evaluating performance on HotpotQA, 2WikiMultiHopQA, and UltraDomain subsets.
Entities (10)
Relation Signals (9)
ACE-GraphRAG → evaluatedon → 2WikiMultiHopQA
confidence 95% · We evaluate ACE-GraphRAG on ... 2WikiMultiHopQA
ACE-GraphRAG → evaluatedon → HotpotQA
confidence 95% · We evaluate ACE-GraphRAG on HotpotQA
Adaptive-ACE → isvariantof → ACE-GraphRAG
confidence 95% · Adaptive-ACE selects task- and topology-specific policies for individual queries
Full-ACE → isvariantof → ACE-GraphRAG
confidence 95% · Full-ACE applies the full policy uniformly within each task family
ACE-GraphRAG → addresses → representation-inference gap
confidence 90% · We identify this mismatch as the representation--inference gap. We propose ... ACE-GraphRAG
ACE-GraphRAG → evaluatedon → UltraDomain
confidence 90% · We evaluate ACE-GraphRAG on ... four UltraDomain subsets
ACE-GraphRAG → includescomponent → Parallel Differential Retrieval
confidence 90% · ACE-GraphRAG formulates context construction as a policy over ... Parallel Differential Retrieval acquires supplementary evidence
Full-ACE → outperforms → RAG and GraphRAG baselines
confidence 85% · Full-ACE outperforms the evaluated RAG and GraphRAG baselines across both task families
Cypher Suggestions (0)
No Cypher suggestions yet.
Abstract
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.
Tags
Links
- Source: https://arxiv.org/abs/2608.01269v2
- Canonical: https://arxiv.org/abs/2608.01269v2
PDF not stored locally. Use the link above to view on the source site.
Full Text
5,189 characters extracted from source content.
Expand or collapse full text
Skip to main content System maintenance August 4th and 5th Learn more × Search Submit Donate Log in Search arXiv Press Enter to search · Advanced search Computer Science > Computation and Language arXiv:2608.01269v2 (cs) This paper has been withdrawn by Ruiying Chen [Submitted on 2 Aug 2026 (v1), last revised 4 Aug 2026 (this version, v2)] Title:ACE-GraphRAG: Agentic Context Engineering for Hierarchical GraphRAG Authors:Yongfeng Huang, Yuren Lai, Ruiying Chen, Haoyu Huang, Mingming Zhao, James Cheng View a PDF of the paper titled ACE-GraphRAG: Agentic Context Engineering for Hierarchical GraphRAG, by Yongfeng Huang and 5 other authors No PDF available, click to view other formats 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. Comments: Withdrawn because the manuscript was posted prematurely before completion of the required internal review and release authorization Subjects: Computation and Language (cs.CL); Artificial Intelligence (cs.AI) Cite as: arXiv:2608.01269 [cs.CL] (or arXiv:2608.01269v2 [cs.CL] for this version) https://doi.org/10.48550/arXiv.2608.01269 Focus to learn more arXiv-issued DOI via DataCite Submission history From: Ruiying Chen [view email] [v1] Sun, 2 Aug 2026 14:22:31 UTC (2,358 KB) [v2] Tue, 4 Aug 2026 03:31:36 UTC (1 KB) (withdrawn) Full-text links: Access Paper: View a PDF of the paper titled ACE-GraphRAG: Agentic Context Engineering for Hierarchical GraphRAG, by Yongfeng Huang and 5 other authorsWithdrawn No license for this version due to withdrawn Current browse context: cs.CL < prev | next > new | recent | 2026-08 Change to browse by: cs cs.AI References & Citations NASA ADSGoogle Scholar Semantic Scholar export BibTeX citation Loading... BibTeX formatted citation × loading... Data provided by: Bookmark Bibliographic Tools Bibliographic and Citation Tools Bibliographic Explorer Toggle Bibliographic Explorer (What is the Explorer?) Connected Papers Toggle Connected Papers (What is Connected Papers?) Litmaps Toggle Litmaps (What is Litmaps?) scite.ai Toggle scite Smart Citations (What are Smart Citations?) Code, Data, Media Code, Data and Media Associated with this Article alphaXiv Toggle alphaXiv (What is alphaXiv?) Links to Code Toggle CatalyzeX Code Finder for Papers (What is CatalyzeX?) DagsHub Toggle DagsHub (What is DagsHub?) GotitPub Toggle Gotit.pub (What is GotitPub?) Huggingface Toggle Hugging Face (What is Huggingface?) ScienceCast Toggle ScienceCast (What is ScienceCast?) Demos Demos Replicate Toggle Replicate (What is Replicate?) Spaces Toggle Hugging Face Spaces (What is Spaces?) Spaces Toggle TXYZ.AI (What is TXYZ.AI?) Related Papers Recommenders and Search Tools Link to Influence Flower Influence Flower (What are Influence Flowers?) Core recommender toggle CORE Recommender (What is CORE?) Author Venue Institution Topic About arXivLabs arXivLabs: experimental projects with community collaborators 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. Which authors of this paper are endorsers? | Disable MathJax (What is MathJax?) mathjaxToggle(); We gratefully acknowledge support from our major funders, member institutions, , and all contributors. About · Help · Contact · Subscribe · Copyright · Privacy · Accessibility · Operational Status (opens in new tab) Major funding support from