Elamur Research⌕
☰
home / papers / 2026 / CIRAG: Construction–Integration Retrieval and Adaptive Generation for Multi-hop Question Answering

CIRAG: Construction–Integration Retrieval and Adaptive Generation for Multi-hop Question Answering

Paper · 2026 · Zili Wei, Yilin Wang, Xiaocui Yang et al.

Summary

Triple-based iterative RAG for multi-hop QA has two weaknesses. It keeps only one best triple per step, so early mistakes compound and parallel evidence is lost. It also uses a single evidence granularity, which can be too noisy or too sparse depending on the question. CIRAG draws on Kintsch's Construction-Integration model from cognitive psychology to address both.

  • ICI (Iterative Construction-Integration): an LLM extracts candidate triples from retrieved documents, and the retriever reranks them. A discriminator then filters them against the original question and the accumulated history, keeps a core triple set and writes the next-hop query. This preserves several evidence chains.
  • ACMG: answers start from triples and escalate to supporting sentences or full passages only when needed.
  • Trajectory Distillation: transfers the teacher's integration behaviour to a small student model.

Retrieval recall and answer quality improve over iRAG baselines such as KiRAG, IRCoT, FLARE, MetaRAG and DualRAG, at moderate latency. This makes structured, noise-controlled retrieval more practical for multi-step reasoning.

Key points

  • Retrieval recall: on 100 manually annotated 2WikiMQA questions, ICI reaches triple-level R@3/R@5 of 71.5/82.3. KiRAG scores 66.3/71.3 and NV-Embed-v2 scores 38.3/49.5. ICI also leads at sentence level (89.0 R@5) and passage level (91.5 R@5).
  • Extra datasets: with Qwen2.5-7B-Instruct on WebQ and NQ, CIRAG has the top F1 on both (61.4 and 48.3). MetaRAG has a slightly higher EM on one of them (47.8 vs 46.0).
  • Iteration depth: performance rises from L = 2 to L = 4 iterations and then plateaus, whereas baselines become unstable as noise accumulates.
  • Efficiency: triples are extracted and cached offline, which greatly cuts latency compared with an online-extraction variant, at negligible accuracy cost.
  • Limitations: prompt-based open IE may miss implicit or domain-specific relations, and cascaded granularity checks add latency. Code is available at github.com/52566rz/CIRAG.

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

🎓Northeastern UniversityShenyang · China42 works · 1,940 citations

 Fields, topics and concepts

📚 Shelves 6The field of science this work belongs to (OpenAlex fields).Computer Science · 16,667 worksArtificial Intelligence · 16,194 worksArts and Humanities · 16,039 worksLanguage and Linguistics · 15,891 worksComputer Vision and Pattern Recognition · 2,869 worksInformation Systems · 1,437 works

🧭 Topics 2Research subjects and languages the work is about.Question answering · 441 worksInformation retrieval & search · 473 works

Cite

Cite as Zili Wei, Yilin Wang, Xiaocui Yang, Shi Feng, Weidong Bao, Daling Wang, Zihan Wang and Yifei Zhang. 2026. CIRAG: Construction–Integration Retrieval and Adaptive Generation for Multi-hop Question Answering. 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.1203

@inproceedings{wei2026cirag,
  title = {{CIRAG: Construction–Integration Retrieval and Adaptive Generation for Multi-hop Question Answering}},
  author = {Zili Wei, Yilin Wang, Xiaocui Yang, Shi Feng, Weidong Bao, Daling Wang, Zihan Wang and Yifei Zhang},
  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.1203},
  url = {https://doi.org/10.18653/v1/2026.acl-long.1203}
}

References

44 references, 27 of them with a card of ours. ↗ leads to the work, «find» searches for one we have no id for yet.

  1. Akari Asai, Zeqiu Wu, Yizhong Wang, Avirup Sil, and Hannaneh Hajishirzi. 2024. Self-RAG: Learning to retrieve, generate, and critique through self-reflection. In The Twelfth International Conference on Learning Representations.
  2. Jonathan Berant, Andrew Chou, Roy Frostig, and Percy Liang. 2013. Semantic parsing on freebase from question-answer pairs. In Proceedings of the 2013 Conference on Empirical Methods in Natural Language Processing, pages 1533–1544.
  3. Baian Chen, Chang Shu, Ehsan Shareghi, Nigel Collier, Karthik Narasimhan, and Shunyu Yao. 2023. Fireact: Toward language agent fine-tuning. arXiv preprint arXiv:2310.05915.
  4. Rong Cheng, Jinyi Liu, Yan Zheng, Fei Ni, Jiazhen Du, Hangyu Mao, Fuzheng Zhang, Bo Wang, and Jianye Hao. 2025. Dualrag: A dual-process approach to integrate reasoning and retrieval for multi-hop question answering. In Proceedings of the 63rd Annual Meeting of the Association for Computational Lingu…
  5. Darren Edge, Ha Trinh, Newman Cheng, Joshua Bradley, Alex Chao, Apurva Mody, Steven Truitt, and Jonathan Larson. 2024. From local to global: A graph RAG approach to query-focused summarization. arXiv preprint arXiv:2404.16130. find
  6. Wenqi Fan, Yujuan Ding, Liangbo Ning, Shijie Wang, Hengyun Li, Dawei Yin, Tat-Seng Chua, and Qing Li. 2024. A survey on rag meeting llms: Towards retrieval-augmented large language models. In Proceedings of the 30th ACM SIGKDD conference on knowledge discovery and data mining, pages 6491– 6501.
  7. Jinyuan Fang, Zaiqiao Meng, and Craig MacDonald. 2024. TRACE the evidence: Constructing knowledge-grounded reasoning chains for retrieval-augmented generation. In Findings of the Association for Computational Linguistics: EMNLP, pages 8472–8494. find
  8. Jinyuan Fang, Zaiqiao Meng, and Craig Macdonald. 2025. Kirag: Knowledge-driven iterative retriever for enhancing retrieval-augmented generation. In Proceedings of the 63rd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers), pages 18969–18985. find
  9. Zhibin Gou, Zhihong Shao, Yeyun Gong, Yelong Shen, Yujiu Yang, Minlie Huang, Nan Duan, and Weizhu Chen. 2023. Tora: A tool-integrated reasoning agent for mathematical problem solving. arXiv preprint arXiv:2309.17452.
  10. Aaron Grattafiori, Abhimanyu Dubey, Abhinav Jauhri, Abhinav Pandey, Abhishek Kadian, Ahmad Al-Dahle, Aiesha Letman, Akhil Mathur, Alan Schelten, Alex Vaughan, and 1 others. 2024. The llama 3 herd of models. arXiv preprint arXiv:2407.21783.
  11. Bernal J Gutiérrez, Yiheng Shu, Yu Gu, Michihiro Yasunaga, and Yu Su. 2024. Hipporag: Neurobiologically inspired long-term memory for large language models. Advances in neural information processing systems, 37:59532–59569.
  12. 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. find
  13. Xanh Ho, Anh-Khoa Duong Nguyen, Saku Sugawara, and Akiko Aizawa. 2020. Constructing A multi-hop QA dataset for comprehensive evaluation of reasoning steps. In Proceedings of the 28th International Conference on Computational Linguistics, pages 6609–6625.
  14. Edward J Hu, Yelong Shen, Phillip Wallis, Zeyuan Allen-Zhu, Yuanzhi Li, Shean Wang, Lu Wang, Weizhu Chen, and 1 others. 2022. Lora: Low-rank adaptation of large language models. ICLR, 1(2):3.
  15. Zhengbao Jiang, Frank F. Xu, Luyu Gao, Zhiqing Sun, Qian Liu, Jane Dwivedi-Yu, Yiming Yang, Jamie Callan, and Graham Neubig. 2023. Active retrieval augmented generation. In Proceedings of the 2023 Conference on Empirical Methods in Natural Language Processing, pages 7969–7992.
  16. Li Jiapeng, Liu Runze, Li Yabo, Zhou Tong, Li Mingling, and Chen Xiang. 2024. Tree of reviews: A tree-based dynamic iterative retrieval framework for multi-hop question answering. arXiv preprint arXiv:2404.14464. — [arXiv:2404.14464](https://arxiv.org/abs/2404.14464) find
  17. Minki Kang, Jongwon Jeong, Seanie Lee, Jaewoong Cho, and Sung Ju Hwang. 2025. Distilling llm agent into small models with retrieval and code tools. arXiv preprint arXiv:2505.17612.
  18. Walter Kintsch and Teun A Van Dijk. 1978. Toward a model of text comprehension and production. Psychological review, 85(5):363. find
  19. Tom Kwiatkowski, Jennimaria Palomaki, Olivia Redfield, Michael Collins, Ankur Parikh, Chris Alberti, Danielle Epstein, Illia Polosukhin, Jacob Devlin, Kenton Lee, and 1 others. 2019. Natural questions: a benchmark for question answering research. Transactions of the Association for Computational Lin…
  20. Chankyu Lee, Rajarshi Roy, Mengyao Xu, Jonathan Raiman, Mohammad Shoeybi, Bryan Catanzaro, and Wei Ping. 2024. Nv-embed: Improved techniques for training llms as generalist embedding models. arXiv preprint arXiv:2405.17428.
  21. Hyunji Lee, Sohee Yang, Hanseok Oh, and Minjoon Seo. 2022. Generative multi-hop retrieval. In Proceedings of the 2022 Conference on Empirical Methods in Natural Language Processing, pages 1417– 1436. find
  22. Patrick Lewis, Ethan Perez, Aleksandra Piktus, Fabio Petroni, Vladimir Karpukhin, Naman Goyal, Heinrich Küttler, Mike Lewis, Wen-tau Yih, Tim Rocktäschel, and 1 others. 2020. Retrieval-augmented generation for knowledge-intensive NLP tasks. Advances in Neural Information Processing Systems, 33:9459–…
  23. Rui Li, Quanyu Dai, Zeyu Zhang, Xu Chen, Zhenhua Dong, and Ji-Rong Wen. 2025. Knowtrace: Bootstrapping iterative retrieval-augmented generation with structured knowledge tracing. In Proceedings of the 31st ACM SIGKDD Conference on Knowledge Discovery and Data Mining V. 2, pages 1470–1480. find
  24. Xi Victoria Lin, Xilun Chen, Mingda Chen, Weijia Shi, Maria Lomeli, Rich James, Pedro Rodriguez, Jacob Kahn, Gergely Szilvasy, Mike Lewis, and 1 others. 2024. RA-DIT: Retrieval-augmented dual instruction tuning. In International Conference on Learning Representations.
  25. Linhao Luo, Yuan-Fang Li, Gholamreza Haffari, and Shirui Pan. 2024. Reasoning on graphs: Faithful and interpretable large language model reasoning. In The Twelfth International Conference on Learning Representations. find
  26. Alex Mallen, Akari Asai, Victor Zhong, Rajarshi Das, Daniel Khashabi, and Hannaneh Hajishirzi. 2023. When not to trust language models: Investigating effectiveness of parametric and non-parametric memories. In Proceedings of the 61st Annual Meeting of the Association for Computational Linguistics (V… find
  27. Ori Ram, Yoav Levine, Itay Dalmedigos, Dor Muhlgay, Amnon Shashua, Kevin Leyton-Brown, and Yoav Shoham. 2023. In-context retrieval-augmented language models. Transactions of the Association for Computational Linguistics, 11:1316–1331.
  28. Matthew Renze. 2024. The effect of sampling temperature on problem solving in large language models. In Findings of the association for computational linguistics: EMNLP 2024, pages 7346–7356.
  29. Zhihong Shao, Yeyun Gong, Yelong Shen, Minlie Huang, Nan Duan, and Weizhu Chen. 2023. Enhancing retrieval-augmented large language models with iterative retrieval-generation synergy. In Findings of the Association for Computational Linguistics: EMNLP, pages 9248–9274.
  30. Freda Shi, Xinyun Chen, Kanishka Misra, Nathan Scales, David Dohan, Ed H Chi, Nathanael Schärli, and Denny Zhou. 2023. Large language models can be easily distracted by irrelevant context. In International Conference on Machine Learning, pages 31210–31227. PMLR.
  31. Cunxiang Wang, Xiaoze Liu, Yuanhao Yue, Xiangru Tang, Tianhang Zhang, Cheng Jiayang, Yunzhi Yao, Wenyang Gao, Xuming Hu, Zehan Qi, and 1 others. 2023. Survey on factuality in large language models: Knowledge, retrieval and domain-specificity. arXiv preprint arXiv:2310.07521.
  32. Jingjin Wang and Jiawei Han. 2025. Proprag: Guiding retrieval with beam search over proposition paths. In Proceedings of the 2025 Conference on Empirical Methods in Natural Language Processing, pages 6223–6238. find
  33. Cathleen Wharton and Walter Kintsch. 1991. An overview of construction-integration model: a theory of comprehension as a foundation for a new cognitive architecture. ACM Sigart Bulletin, 2(4):169– 173. find
  34. Shitao Xiao, Zheng Liu, Peitian Zhang, Niklas Muennighoff, Defu Lian, and Jian-Yun Nie. 2024. C-pack: Packed resources for general chinese embeddings. In Proceedings of the 47th international ACM SIGIR conference on research and development in information retrieval, pages 641–649.
  35. An Yang, Baosong Yang, Beichen Zhang, Binyuan Hui, Bo Zheng, Bowen Yu, Chengyuan Li, Dayiheng Liu, Fei Huang, Haoran Wei, and 1 others. 2024. Qwen2. 5 technical report. arXiv preprint arXiv:2412.15115. find
  36. 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. In Proceedings of the 2018 Conference on Empirical Methods in Natural Language Processing, pages 2…
  37. Zijun Yao, Weijian Qi, Liangming Pan, Shulin Cao, Linmei Hu, Liu Weichuan, Lei Hou, and Juanzi Li. 2025. Seakr: Self-aware knowledge retrieval for adaptive retrieval augmented generation. In Proceedings of the 63rd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Paper…
  38. Ori Yoran, Tomer Wolfson, Ori Ram, and Jonathan Berant. 2024. Making retrieval-augmented language models robust to irrelevant context. In International Conference on Learning Representations.
  39. Harsh Trivedi, Niranjan Balasubramanian, Tushar Khot, and Ashish Sabharwal. 2022. MuSiQue: Multihop questions via single-hop question composition. Trans. Assoc. Comput. Linguistics, 10:539–554.
  40. Chao Zhang, Yuhao Wang, Derong Xu, Haoxin Zhang, Yuanjie Lyu, Yuhao Chen, Shuochen Liu, Tong Xu, Xiangyu Zhao, Yan Gao, and 1 others. 2025. Tearag: A token-efficient agentic retrieval-augmented generation framework. arXiv preprint arXiv:2511.05385. — [arXiv:2511.05385](https://arxiv.org/abs/2511.05385) find
  41. Harsh Trivedi, Niranjan Balasubramanian, Tushar Khot, and Ashish Sabharwal. 2023. Interleaving retrieval with chain-of-thought reasoning for knowledge-intensive multi-step questions. In Proceedings of the 61st Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers), p…
  42. Jiahao Zhang, Haiyang Zhang, Dongmei Zhang, Liu Yong, and Shen Huang. 2024. End-to-end beam retrieval for multi-hop question answering. In Proceedings of the 2024 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies (Volume 1: Long Pa… find
  43. Chen Zhao, Chenyan Xiong, Jordan L. Boyd-Graber, and Hal Daumé III. 2021. Multi-step reasoning over unstructured text with beam dense retrieval. In Proceedings of the 2021 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, pages 46… find
  44. Yujia Zhou, Zheng Liu, Jiajie Jin, Jian-Yun Nie, and Zhicheng Dou. 2024. Metacognitive retrieval-augmented large language models. In Proceedings of the ACM Web Conference 2024, pages 1453– 1463. find

Updated

↑ back to top

This page in Markdown → https://md.research.elamur.ai/free/en/papers/wei-2026-cirag-construction-integration-retrieval-adaptive/