Elamur Research⌕
☰
home / papers / 2025 / RichRAG: Crafting Rich Responses for Multi-faceted Queries in Retrieval-Augmented Generation

RichRAG: Crafting Rich Responses for Multi-faceted Queries in Retrieval-Augmented Generation

Paper · 2025 · Shuting Wang, Xin Yu, Mang Wang et al.

Summary

Most retrieval-augmented generation (RAG) research targets questions with a clear intent and a short answer. Many real user queries are broad, though, and call for long responses that cover several hidden sub-topics. RichRAG addresses this with a three-part pipeline. A fine-tuned LLM first predicts the sub-aspects a query may involve. A retriever then fetches documents for each sub-aspect, combined with the original query, and merges them into a diverse candidate pool. Finally, a generative list-wise ranker selects the top-k documents passed to a fixed generator LLM. The ranker uses an encoder-decoder (FiD-style) design that reads all candidates together and outputs document IDs. It is trained in two stages. Supervised fine-tuning uses greedy, coverage-based silver rankings. DPO then aligns rankings with what helps the generator produce accurate, comprehensive answers. On two public datasets, RichRAG yields richer answers than existing approaches while staying efficient. This shows that LLM-aware, coverage-oriented ranking matters for open-ended queries.

Key points

  • The framework has three modules: a sub-aspect explorer (an LLM fine-tuned with next-token prediction to output bracketed sub-aspects), a multi-faceted retriever, and a generative list-wise ranker.
  • The ranker follows the FiD architecture. It encodes candidates in parallel and adds pooling and reuse in the decoder to reduce time and memory costs.
  • The US³ (unilateral significance sampling) strategy builds DPO training pairs. Ablations show that random pairs, removing RL ("w/o RL"), or removing sub-aspects or the global ranker all hurt performance.
  • In efficiency tests against LDIST (point-wise) and LiT5 (sliding-window list-wise), RichRAG's latency grows like the point-wise method. LiT5's latency rises much more steeply as candidates increase. Across pools of 50–200 candidates, the latency gap stays within about 1 second.
  • The authors note two limitations: predicted sub-aspects only partly reflect real user intents, and few suitable datasets exist (evaluation includes WikipassageQA).

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

🎓Renmin University of ChinaBeijing · China139 works · 17,752 citations🏢BaichuanBeijing · China8 works · 317 citations

 Fields, topics and concepts

📚 Shelves 6The 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 worksComputer Vision and Pattern Recognition · 3,486 worksInformation Systems · 1,730 works

🧭 Topics 1Research subjects and languages the work is about.Information retrieval & search · 552 works

Cite

Cite as Shuting Wang, Xin Yu, Mang Wang, Weipeng Chen, Yutao Zhu and Zhicheng Dou. 2025. RichRAG: Crafting Rich Responses for Multi-faceted Queries in Retrieval-Augmented Generation. Proceedings of the 31st International Conference on Computational Linguistics.

The paper itself, not this page. BibTeX from the library's record of the work.

@inproceedings{wang2025richrag,
  title = {{RichRAG: Crafting Rich Responses for Multi-faceted Queries in Retrieval-Augmented Generation}},
  author = {Shuting Wang, Xin Yu, Mang Wang, Weipeng Chen, Yutao Zhu and Zhicheng Dou},
  booktitle = {Proceedings of the 31st International Conference on Computational Linguistics},
  year = {2025},
  url = {https://aclanthology.org/2025.coling-main.750/}
}

References

43 references, 22 of them with a card of ours. ↗ leads to the work, «find» searches for one we have no id for yet.

  1. Daman Arora, Anush Kini, Sayak Ray Chowdhury, Nagarajan Natarajan, Gaurav Sinha, and Amit Sharma. 2023. Gar-meets-rag paradigm for zero-shot information retrieval. ArXiv, abs/2310.20158. — [arXiv:2310.20158](https://arxiv.org/abs/2310.20158) find
  2. 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.
  3. Sebastian Borgeaud, Arthur Mensch, Jordan Hoffmann, Trevor Cai, Eliza Rutherford, Katie Millican, George Bm Van Den Driessche, Jean-Baptiste Lespiau, Bogdan Damoc, Aidan Clark, Diego De Las Casas, Aurelia Guy, Jacob Menick, Roman Ring, Tom Hennigan, Saffron Huang, Loren Maggiore, Chris Jones, Albin… find
  4. Chi-Min Chan, Chunpu Xu, Ruibin Yuan, Hongyin Luo, Wei Xue, Yike Guo, and Jie Fu. 2024. Rq-rag: Learning to refine queries for retrieval augmented generation. Preprint, arXiv:2404.00610.
  5. Hyung Won Chung, Le Hou, Shayne Longpre, Barret Zoph, Yi Tay, William Fedus, Eric Li, Xuezhi Wang, Mostafa Dehghani, Siddhartha Brahma, Albert Webson, Shixiang Shane Gu, Zhuyun Dai, Mirac Suzgun, Xinyun Chen, Aakanksha Chowdhery, Sharan Narang, Gaurav Mishra, Adams Yu, Vincent Zhao, Yanping Huang, A… find
  6. Gordon V. Cormack, Charles L A Clarke, and Stefan Buettcher. 2009. Reciprocal rank fusion outperforms condorcet and individual rank learning methods. In Proceedings of the 32nd International ACM SIGIR Conference on Research and Development in Information Retrieval, SIGIR ’09, page 758–759, New York,…
  7. Van Dang and W. Bruce Croft. 2012. Diversity by proportionality: an election-based approach to search result diversification. In Proceedings of the 35th International ACM SIGIR Conference on Research and Development in Information Retrieval, SIGIR ’12, page 65–74, New York, NY, USA. Association for… find
  8. 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.
  9. Hiroaki Hayashi, Prashant Budania, Peng Wang, Chris Ackerson, Raj Neervannan, and Graham Neubig. 2021. WikiAsp: A dataset for multi-domain aspect-based summarization. Transactions of the Association for Computational Linguistics, 9:211–225.
  10. Gautier Izacard and Edouard Grave. 2021. Leveraging passage retrieval with generative models for open domain question answering. In Proceedings of the 16th Conference of the European Chapter of the Association for Computational Linguistics: Main Volume, pages 874–880, Online. Association for Computa…
  11. Gautier Izacard, Patrick Lewis, Maria Lomeli, Lucas Hosseini, Fabio Petroni, Timo Schick, Jane Dwivedi-Yu, Armand Joulin, Sebastian Riedel, and Edouard Grave. 2024. Atlas: few-shot learning with retrieval augmented language models. J. Mach. Learn. Res., 24(1).
  12. Zhengbao Jiang, Frank 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, Singapore. Association fo…
  13. Zixuan Ke, Weize Kong, Cheng Li, Mingyang Zhang, Qiaozhu Mei, and Michael Bendersky. 2024. Bridging the preference gap between retrievers and llms. Preprint, arXiv:2401.06954.
  14. Tushar Khot, Harsh Trivedi, Matthew Finlayson, Yao Fu, Kyle Richardson, Peter Clark, and Ashish Sabharwal. 2023. Decomposed prompting: A modular approach for solving complex tasks. Preprint, arXiv:2210.02406.
  15. 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. 2020. Retrieval-augmented generation for knowledge-intensive nlp tasks. In Proceedings of the 34th Internation…
  16. Xinze Li, Zhenghao Liu, Chenyan Xiong, Shi Yu, Yukun Yan, Shuo Wang, and Ge Yu. 2024. Say more with less: Understanding prompt learning behaviors through gist compression. Preprint, arXiv:2402.16058. — [arXiv:2402.16058](https://arxiv.org/abs/2402.16058) find
  17. Xi Victoria Lin, Xilun Chen, Mingda Chen, Weijia Shi, Maria Lomeli, Rich James, Pedro Rodriguez, Jacob Kahn, Gergely Szilvasy, Mike Lewis, Luke Zettlemoyer, and Scott Yih. 2023. Ra-dit: Retrieval-augmented dual instruction tuning. ArXiv, abs/2310.01352.
  18. Jiongnan Liu, Zhicheng Dou, Xiaojie Wang, Shuqi Lu, and Ji-Rong Wen. 2020. Dvgan: A minimax game for search result diversification combining explicit and implicit features. In Proceedings of the 43rd International ACM SIGIR Conference on Research and Development in Information Retrieval, SIGIR ’20,… find
  19. Ilya Loshchilov and Frank Hutter. 2019. Decoupled weight decay regularization. In 7th International Conference on Learning Representations, ICLR 2019, New Orleans, LA, USA, May 6-9, 2019. find
  20. Rodrigo Nogueira, Zhiying Jiang, Ronak Pradeep, and Jimmy Lin. 2020. Document ranking with a pre-trained sequence-to-sequence model. In Findings of the Association for Computational Linguistics: EMNLP 2020, pages 708–718, Online. Association for Computational Linguistics. find
  21. Ashwin Paranjape, Omar Khattab, Christopher Potts, Matei Zaharia, and Christopher D. Manning. 2022. Hindsight: Posterior-guided training of retrievers for improved open-ended generation. In The Tenth International Conference on Learning Representations, ICLR 2022, Virtual Event, April 25-29, 2022. O… find
  22. Zhen Qin, Rolf Jagerman, Kai Hui, Honglei Zhuang, Junru Wu, Le Yan, Jiaming Shen, Tianqi Liu, Jialu Liu, Donald Metzler, Xuanhui Wang, and Michael Bendersky. 2024. Large language models are effective text rankers with pairwise ranking prompting. Preprint, arXiv:2306.17563.
  23. Zackary Rackauckas. 2024. Rag-fusion: A new take on retrieval augmented generation. International Journal on Natural Language Computing, 13(1):37–47. find
  24. Rafael Rafailov, Archit Sharma, Eric Mitchell, Christopher D Manning, Stefano Ermon, and Chelsea Finn. 2023. Direct preference optimization: Your language model is secretly a reward model. In Thirty-seventh Conference on Neural Information Processing Systems.
  25. Colin Raffel, Noam Shazeer, Adam Roberts, Katherine Lee, Sharan Narang, Michael Matena, Yanqi Zhou, Wei Li, and Peter J. Liu. 2020. Exploring the limits of transfer learning with a unified text-to-text transformer. Journal of Machine Learning Research, 21(140):1–67.
  26. Rodrygo L.T. Santos, Craig Macdonald, and Iadh Ounis. 2010. Exploiting query reformulations for web search result diversification. In Proceedings of the 19th International Conference on World Wide Web, WWW ’10, page 881–890, New York, NY, USA. Association for Computing Machinery. find
  27. Weijia Shi, Sewon Min, Michihiro Yasunaga, Minjoon Seo, Rich James, Mike Lewis, Luke Zettlemoyer, and Wen tau Yih. 2023. Replug: Retrieval-augmented black-box language models. Preprint, arXiv:2301.12652.
  28. Weiwei Sun, Lingyong Yan, Xinyu Ma, Shuaiqiang Wang, Pengjie Ren, Zhumin Chen, Dawei Yin, and Zhaochun Ren. 2023. Is ChatGPT good at search? investigating large language models as re-ranking agents. In Proceedings of the 2023 Conference on Empirical Methods in Natural Language Processing, pages 1491…
  29. Manveer Singh Tamber, Ronak Pradeep, and Jimmy Lin. 2023. Scaling down, litting up: Efficient zero-shot listwise reranking with seq2seq encoder-decoder models. Preprint, arXiv:2312.16098. — [arXiv:2312.16098](https://arxiv.org/abs/2312.16098) find
  30. Zhiruo Wang, Jun Araki, Zhengbao Jiang, Md Rizwan Parvez, and Graham Neubig. 2023d. Learning to filter context for retrieval-augmented generation. Preprint, arXiv:2311.08377.
  31. Hugo Touvron, Louis Martin, Kevin Stone, Peter Albert, Amjad Almahairi, Yasmine Babaei, Nikolay Bashlykov, Soumya Batra, Prajjwal Bhargava, Shruti Bhosale, Dan Bikel, Lukas Blecher, Cristian Canton Ferrer, Moya Chen, Guillem Cucurull, David Esiobu, Jude Fernandes, Jeremy Fu, Wenyin Fu, Brian Fuller,… — Llama 2: Open Foundation and Fine-Tuned Chat Models
  32. Shitao Xiao, Zheng Liu, Peitian Zhang, and Niklas Muennighoff. 2023. C-pack: Packaged resources to advance general chinese embedding. Preprint, arXiv:2309.07597.
  33. Boxin Wang, Wei Ping, Peng Xu, Lawrence McAfee, Zihan Liu, Mohammad Shoeybi, Yi Dong, Oleksii Kuchaiev, Bo Li, Chaowei Xiao, Anima Anandkumar, and Bryan Catanzaro. 2023a. Shall we pretrain autoregressive language models with retrieval? a comprehensive study. In Proceedings of the 2023 Conference on… find
  34. Hongru Wang, Boyang Xue, Baohang Zhou, Tianhua Zhang, Cunxiang Wang, Guanhua Chen, Huimin Wang, and Kam fai Wong. 2024a. Self-dc: When to retrieve and when to generate? self divide-and-conquer for compositional unknown questions. Preprint, arXiv:2402.13514. — [arXiv:2402.13514](https://arxiv.org/abs/2402.13514) find
  35. Shuting Wang, Zhicheng Dou, Jing Yao, Yujia Zhou, and Ji-Rong Wen. 2023b. Incorporating explicit subtopics in personalized search. In Proceedings of the ACM Web Conference 2023, WWW ’23, page 3364–3374, New York, NY, USA. Association for Computing Machinery. find
  36. Shuting Wang, Jiongnan Liu Shiren Song, Jiehan Cheng, Yuqi Fu, Peidong Guo, Kun Fang, Yutao Zhu, and Zhicheng Dou. 2024b. Domainrag: A chinese benchmark for evaluating domain-specific retrieval-augmented generation. Preprint, arXiv:2406.05654. — [arXiv:2406.05654](https://arxiv.org/abs/2406.05654) find
  37. Fangyuan Xu, Weijia Shi, and Eunsol Choi. 2023. Recomp: Improving retrieval-augmented lms with compression and selective augmentation. Preprint, arXiv:2310.04408.
  38. Shicheng Xu, Liang Pang, Huawei Shen, Xueqi Cheng, and Tat-Seng Chua. 2024a. Search-in-the-chain: Interactively enhancing large language models with search for knowledge-intensive tasks. Preprint, arXiv:2304.14732. — [arXiv:2304.14732](https://arxiv.org/abs/2304.14732) find
  39. Shicheng Xu, Liang Pang, Jun Xu, Huawei Shen, and Xueqi Cheng. 2024b. List-aware reranking-truncation joint model for search and retrieval-augmented generation. Preprint, arXiv:2402.02764. — [arXiv:2402.02764](https://arxiv.org/abs/2402.02764) find
  40. Shunyu Yao, Jeffrey Zhao, Dian Yu, Nan Du, Izhak Shafran, Karthik Narasimhan, and Yuan Cao. 2023. React: Synergizing reasoning and acting in language models. Preprint, arXiv:2210.03629.
  41. Zichun Yu, Chenyan Xiong, Shi Yu, and Zhiyuan Liu. 2023. Augmentation-adapted retriever improves generalization of language models as generic plug-in. In Proceedings of the 61st Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers), pages 2421–2436, Toronto, Canada.… find
  42. Honglei Zhuang, Zhen Qin, Rolf Jagerman, Kai Hui, Ji Ma, Jing Lu, Jianmo Ni, Xuanhui Wang, and Michael Bendersky. 2023a. Rankt5: Fine-tuning t5 for text ranking with ranking losses. In Proceedings of the 46th International ACM SIGIR Conference on Research and Development in Information Retrieval, SI… find
  43. Shengyao Zhuang, Honglei Zhuang, Bevan Koopman, and Guido Zuccon. 2023b. A setwise approach for effective and highly efficient zero-shot ranking with large language models. Preprint, arXiv:2310.09497. — [arXiv:2310.09497](https://arxiv.org/abs/2310.09497) find

Updated

↑ back to top

This page in Markdown → https://md.research.elamur.ai/free/en/papers/wang-2025-richrag-crafting-rich-responses-multi/