Elamur Research⌕
☰
home / papers / 2026 / Query-Aware Knowledge Retrieval via Hyperbolic Structuring

Query-Aware Knowledge Retrieval via Hyperbolic Structuring

Paper · 2026 · Chuang Zhou, Junnan Dong, Yilin Xiao et al.

Summary

Graph-based retrieval-augmented generation (GraphRAG) usually builds one fixed graph over a corpus, using heuristics like entity co-occurrence or clustering. Such a graph ignores the particular reasoning chain a given question needs. HyperRAG addresses this by building a graph adapted to each query. It first links passages that share enough key terms, giving a static, corpus-level backbone. An LLM then breaks the query into atomic reasoning units, and each unit retrieves its own evidence. This creates implicit, query-specific links between passages. The two kinds of connections are merged into a single cycle-free tree by unifying shared nodes. Passages are then embedded in a Poincaré ball, a hyperbolic space whose exponential volume growth suits tree structures and keeps passages many reasoning hops apart well separated. The authors report that HyperRAG outperforms existing baselines on three benchmark datasets without notable extra computational cost. They also find that query-induced edges contribute more than co-occurrence edges. The work suggests that query-adaptive structure and non-Euclidean geometry can improve multi-hop retrieval.

Key points

  • Explicit edges: passages are linked when their key-term overlap ratio reaches θ = 0.15. Each passage is capped at five such neighbours.
  • Implicit edges: the LLM decomposes each query into minimal reasoning units, and each unit triggers a targeted retrieval. The resulting query-specific subtrees are fused into one global tree, with loop-creating edges removed.
  • Ablation: co-occurrence edges give a small gain on all datasets, while query-guided implicit edges improve results substantially.
  • Distance separation: hyperbolic distances spread more widely than Euclidean ones, avoiding the distance concentration seen in high dimensions. Mean query–passage distances for Top-1/3/10 were 1.87/1.97/2.10 in hyperbolic space versus 0.71/0.87/1.06 in Euclidean space.
  • Limitation: hyperbolic search cannot use optimized libraries like FAISS, so retrieval is somewhat slower. The authors plan parallel and multi-process search to address this.

Our summary of the paper, in our own words — not its text. The work belongs to its authors; read it at the links above.

 Institutions

🏢TencentShenzhen · China163 works · 21,032 citations🎓Hong Kong Polytechnic UniversityHong Kong · China106 works · 5,190 citations🎓Stevens Institute of TechnologyHoboken · United States13 works · 95 citations

 Fields, topics and concepts

📚 Shelves 5The field of science this work belongs to (OpenAlex fields).Computer Science · 18,614 worksArtificial Intelligence · 17,965 worksArts and Humanities · 17,450 worksLanguage and Linguistics · 17,296 worksInformation Systems · 1,730 works

🧭 Topics 2Research subjects and languages the work is about.Reasoning · 1,685 worksInformation retrieval & search · 552 works

Cite

Cite as Chuang Zhou, Junnan Dong, Yilin Xiao, Shengyuan Chen, Su Dong, di Yin, Xing Sun, Zhaozhuo Xu and Xiao Huang. 2026. Query-Aware Knowledge Retrieval via Hyperbolic Structuring. Proceedings of the 64th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers).

The paper itself, not this page. BibTeX from the library's record of the work. DOI 10.18653/v1/2026.acl-long.986

@inproceedings{zhou2026query,
  title = {{Query-Aware Knowledge Retrieval via Hyperbolic Structuring}},
  author = {Chuang Zhou, Junnan Dong, Yilin Xiao, Shengyuan Chen, Su Dong, di Yin, Xing Sun, Zhaozhuo Xu and Xiao Huang},
  booktitle = {Proceedings of the 64th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)},
  year = {2026},
  doi = {10.18653/v1/2026.acl-long.986},
  url = {https://doi.org/10.18653/v1/2026.acl-long.986}
}

References

40 references, 18 of them with a card of ours. ↗ leads to the work, «find» searches for one we have no id for yet.

  1. Josh Achiam, Steven Adler, Sandhini Agarwal, Lama Ahmad, Ilge Akkaya, Florencia Leoni Aleman, Diogo Almeida, Janko Altenschmidt, Sam Altman, Shyamal Anadkat, and 1 others. 2023. Gpt-4 technical report. arXiv preprint arXiv:2303.08774.
  2. Payal Bajaj, Chenyan Xiong, Guolin Ke, Xiaodong Liu, Di He, Saurabh Tiwary, Tie-Yan Liu, Paul Bennett, Xia Song, and Jianfeng Gao. 2022. Metro: Efficient denoising pretraining of large scale autoencoding language models with model generated signals. arXiv preprint arXiv:2204.06644. — [arXiv:2204.06644](https://arxiv.org/abs/2204.06644) find
  3. Shengyuan Chen, Chuang Zhou, Zheng Yuan, Qinggang Zhang, Zeyang Cui, Hao Chen, Yilin Xiao, Jiannong Cao, and Xiao Huang. 2026. You don’t need pre-built graphs for rag: Retrieval augmented generation with adaptive reasoning structures. In Proceedings of the AAAI Conference on Artificial Intelligence,…
  4. Roland Coghetto. 2016. Chebyshev distance. find
  5. HWJ Debeye and P Van Riel. 1990. Lp-norm deconvolution1. Geophysical Prospecting, 38(4):381–403. find
  6. Junnan Dong, Siyu An, Yifei Yu, Qian-Wen Zhang, Linhao Luo, Xiao Huang, Yunsheng Wu, Di Yin, and Xing Sun. 2025. Youtu-graphrag: Vertically unified agents for graph retrieval-augmented complex reasoning. arXiv preprint arXiv:2508.19855. — [arXiv:2508.19855](https://arxiv.org/abs/2508.19855) find
  7. Junnan Dong, Qinggang Zhang, Xiao Huang, Keyu Duan, Qiaoyu Tan, and Zhimeng Jiang. 2023. Hierarchy-aware multi-hop question answering over knowledge graphs. In Proceedings of the ACM web conference 2023, pages 2519–2527. find
  8. Darren Edge, Ha Trinh, Newman Cheng, Joshua Bradley, Alex Chao, Apurva Mody, Steven Truitt, Dasha Metropolitansky, Robert Osazuwa Ness, and Jonathan Larson. 2024. From local to global: A graph rag approach to query-focused summarization. arXiv preprint arXiv:2404.16130. find
  9. Damien François, Vincent Wertz, and Michel Verleysen. 2007. The concentration of fractional distances. IEEE Transactions on Knowledge and Data Engineering, 19(7):873–886. find
  10. Yunfan Gao, Yun Xiong, Xinyu Gao, Kangxiang Jia, Jinliu Pan, Yuxi Bi, Yi Dai, Jiawei Sun, Meng Wang, and Haofen Wang. 2024. Retrieval-augmented generation for large language models: A survey. Preprint, arXiv:2312.10997.
  11. Zirui Guo, Lianghao Xia, Yanhua Yu, Tu Ao, and Chao Huang. 2024. Lightrag: Simple and fast retrieval-augmented generation.
  12. Bernal Jiménez Gutiérrez, Yiheng Shu, Yu Gu, Michihiro Yasunaga, and Yu Su. 2024. Hipporag: Neurobiologically inspired long-term memory for large language models. In The Thirty-eighth Annual Conference on Neural Information Processing Systems.
  13. Bernal Jiménez Gutiérrez, Yiheng Shu, Weijian Qi, Sizhe Zhou, and Yu Su. 2025. From rag to memory: Non-parametric continual learning for large language models. arXiv preprint arXiv:2502.14802.
  14. Kelvin Guu, Kenton Lee, Zora Tung, Panupong Pasupat, and Mingwei Chang. 2020. Retrieval augmented language model pre-training. In International conference on machine learning, pages 3929–3938. PMLR.
  15. Xiaoxin He, Yijun Tian, Yifei Sun, Nitesh Chawla, Thomas Laurent, Yann LeCun, Xavier Bresson, and Bryan Hooi. 2024. G-retriever: Retrieval-augmented generation for textual graph understanding and question answering. Advances in Neural Information Processing Systems, 37:132876–132907.
  16. Xanh Ho, Anh-Khoa Duong Nguyen, Saku Sugawara, and Akiko Aizawa. 2020. Constructing a multi-hop qa dataset for comprehensive evaluation of reasoning steps. arXiv preprint arXiv:2011.01060.
  17. Sebastian Hofstätter, Jiecao Chen, Karthik Raman, and Hamed Zamani. 2023. Fid-light: Efficient and effective retrieval-augmented text generation. In Proceedings of the 46th International ACM SIGIR Conference on Research and Development in Information Retrieval, pages 1437–1447. find
  18. Bernard S Kay. 1992. The principle of locality and quantum field theory on (non globally hyperbolic) curved spacetimes. Reviews in Mathematical Physics, 4(spec01):167–195. find
  19. Marius Kloft, Ulf Brefeld, Sören Sonnenburg, and Alexander Zien. 2011. Lp-norm multiple kernel learning. The Journal of Machine Learning Research, 12:953–997. find
  20. Patrick Lewis, Ethan Perez, Aleksandra Piktus, Fabio Petroni, Vladimir Karpukhin, Naman Goyal, Heinrich Küttler, Mike Lewis, Wen-tau Yih, Tim Rocktäschel, Sebastian Riedel, and Douwe Kiela. 2020a. find
  21. Retrieval-augmented generation for knowledge-intensive nlp tasks. In Advances in Neural Information Processing Systems, volume 33, pages 9459– 9474. Curran Associates, Inc. Patrick Lewis, Ethan Perez, Aleksandra Piktus, Fabio Petroni, Vladimir Karpukhin, Naman Goyal, Heinrich Küttler, Mike Lewis, We…
  22. Dawei Li, Shu Yang, Zhen Tan, Jae Young Baik, Sukwon Yun, Joseph Lee, Aaron Chacko, Bojian Hou, Duy Duong-Tran, Ying Ding, and 1 others. 2024. Dalk: Dynamic co-augmentation of llms and kg to answer alzheimer’s disease questions with scientific literature. arXiv preprint arXiv:2405.04819. find
  23. Xun Liang, Simin Niu, Sensen Zhang, Shichao Song, Hanyu Wang, Jiawei Yang, Feiyu Xiong, Bo Tang, Chenyang Xi, and 1 others. 2024. Empowering large language models to set up a knowledge retrieval indexer via self-learning. arXiv preprint arXiv:2405.16933. — [arXiv:2405.16933](https://arxiv.org/abs/2405.16933) find
  24. Linhao Luo, Zicheng Zhao, Gholamreza Haffari, Dinh Phung, Chen Gong, and Shirui Pan. 2025. Gfm-rag: Graph foundation model for retrieval augmented generation. arXiv preprint arXiv:2502.01113. — [arXiv:2502.01113](https://arxiv.org/abs/2502.01113) find
  25. Maximillian Nickel and Douwe Kiela. 2017. Poincaré embeddings for learning hierarchical representations. Advances in neural information processing systems, 30.
  26. Érick O Rodrigues. 2018. Combining minkowski and chebyshev: New distance proposal and survey of distance metrics using k-nearest neighbours classifier. Pattern Recognition Letters, 110:66–71. find
  27. Devendra Singh Sachan, Mostofa Patwary, Mohammad Shoeybi, Neel Kant, Wei Ping, William L Hamilton, and Bryan Catanzaro. 2021. End-to-end training of neural retrievers for open-domain question answering. arXiv preprint arXiv:2101.00408.
  28. Parth Sarthi, Salman Abdullah, Aditi Tuli, Shubh Khanna, Anna Goldie, and Christopher D Manning. 2024. Raptor: Recursive abstractive processing for tree-organized retrieval. In The Twelfth International Conference on Learning Representations.
  29. Hugo Touvron, Thibaut Lavril, Gautier Izacard, Xavier Martinet, Marie-Anne Lachaux, Timothée Lacroix, Baptiste Rozière, Naman Goyal, Eric Hambro, Faisal Azhar, and 1 others. 2023. Llama: Open and efficient foundation language models. arXiv preprint arXiv:2302.13971.
  30. Harsh Trivedi, Niranjan Balasubramanian, Tushar Khot, and Ashish Sabharwal. 2022. 9835 musique: Multi-hop questions via single-hop question composition. Trans. Assoc. Comput. Linguistics, 10:539–554. find
  31. Yu Wang, Nedim Lipka, Ryan A Rossi, Alexa Siu, Ruiyi Zhang, and Tyler Derr. 2024. Knowledge graph prompting for multi-document question answering. In Proceedings of the AAAI Conference on Artificial Intelligence, volume 38, pages 19206–19214. find
  32. Yilin Xiao, Junnan Dong, Chuang Zhou, Su Dong, Qianwen Zhang, Di Yin, Xing Sun, and Xiao Huang. 2025. Graphrag-bench: Challenging domain-specific reasoning for evaluating graph retrieval-augmented generation. arXiv preprint arXiv:2506.02404. — [arXiv:2506.02404](https://arxiv.org/abs/2506.02404) find
  33. Yilin Xiao, Chuang Zhou, Qinggang Zhang, Bo Li, Qing Li, and Xiao Huang. 2026. Reliable reasoning path: Distilling effective guidance for llm reasoning with knowledge graphs. IEEE Transactions on Knowledge and Data Engineering. find
  34. Chang Yang, Chuang Zhou, Yilin Xiao, Su Dong, Luyao Zhuang, Yujing Zhang, Zhu Wang, Zijin Hong, Zheng Yuan, Zhishang Xiang, and 1 others. 2026. Graph-based agent memory: Taxonomy, techniques, and applications. arXiv preprint arXiv:2602.05665. — [arXiv:2602.05665](https://arxiv.org/abs/2602.05665) find
  35. Zhilin Yang, Peng Qi, Saizheng Zhang, Yoshua Bengio, William W Cohen, Ruslan Salakhutdinov, and Christopher D Manning. 2018. Hotpotqa: A dataset for diverse, explainable multi-hop question answering. arXiv preprint arXiv:1809.09600.
  36. Qinggang Zhang, Shengyuan Chen, Yuanchen Bei, Zheng Yuan, Huachi Zhou, Zijin Hong, Junnan Dong, Hao Chen, Yi Chang, and Xiao Huang. 2025. A survey of graph retrieval-augmented generation for customized large language models. arXiv preprint arXiv:2501.13958.
  37. Chuang Zhou, Junnan Dong, Xiao Huang, Zirui Liu, Kaixiong Zhou, and Zhaozhuo Xu. 2024. Quest: Efficient extreme multi-label text classification with large language models on commodity hardware. In Findings of the Association for Computational Linguistics: EMNLP 2024, pages 3929–3940. find
  38. Chuang Zhou, Jiahe Du, Huachi Zhou, Hao Chen, Feiran Huang, and Xiao Huang. 2025a. Text-attributed graph learning with coupled augmentations. In Proceedings of the 31st International Conference on Computational Linguistics, pages 10865–10876.
  39. Chuang Zhou, Zhu Wang, Shengyuan Chen, Jiahe Du, Qiyuan Zheng, Zhaozhuo Xu, and Xiao Huang. 2025b. Taming language models for text-attributed graph learning with decoupled aggregation. In Proceedings of the 63rd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers),…
  40. Huachi Zhou, Jiahe Du, Chuang Zhou, Chang Yang, Yilin Xiao, Yuxuan Xie, and Xiao Huang. 2025c. Each graph is a new language: Graph learning with llms. In Findings of the Association for Computational Linguistics: ACL 2025, pages 17548–17559. find

Updated

↑ back to top

This page in Markdown → https://md.research.elamur.ai/free/en/papers/zhou-2026-query-aware-knowledge-retrieval-hyperbolic/