Elamur Research⌕
☰
home / papers / 2025 / RaDeR: Reasoning-aware Dense Retrieval Models

RaDeR: Reasoning-aware Dense Retrieval Models

Paper · 2025 · Debrup Das, Sam O'Nuallain and Razieh Rahimi

Summary

Standard retrievers that augment LLMs rely mostly on lexical or semantic matching. They struggle when a relevant document, such as a theorem needed to solve a problem, shares no wording with the query. Existing fixes either interleave LLM reasoning with fixed retrievers or rerank candidates, so they remain bound by the first-stage results. RaDeR addresses this with first-stage dense retrievers, plus a pointwise cross-encoder reranker, trained to judge relevance through reasoning.

The training data comes from math datasets that contain only questions and gold answers. An LLM solves problems with retrieval-augmented Monte Carlo Tree Search and can fetch theorems for subproblems. Theorems on trajectories that reach the correct answer, confirmed by self-reflective checking, become positives. Rejected theorems serve as hard negatives. Queries take several forms: the question itself, CoT steps, LLM-written queries and lexical queries.

Although trained only on math, RaDeR generalizes across BRIGHT and RAR-b while using a small fraction of competitors' training data. This suggests that carefully verified synthetic data matters more than sheer volume for reasoning-intensive retrieval.

Key points

  • Training data is drawn from MATH and NuminaMath. The full set has 43,120 samples, roughly 2.5% of the 1,729,368 used by the concurrent REASONIR. RaDeR still beats REASONIR by 1.1 nDCG@10, a 4.5% relative gain.
  • On BRIGHT, RaDeR exceeds strong baselines by at least 2 nDCG points. Relative nDCG@10 gains reach 37–40% on the TheoremQA questions split and 8–26% on LeetCode.
  • It is reported as the first dense retriever to surpass BM25 in a zero-shot setting where CoT reasoning steps serve as queries.
  • On RAR-b, the MMTEB reasoning subset, it outperforms sparse and open-source models and roughly matches OpenAI-3-large. Models are built on Qwen2.5-7B-instruct and gte-Qwen2-7B-instruct, and code, data and models are released on GitHub and Hugging Face (Raderspace).
  • The authors note three limitations. Relevance is judged one document at a time, scores come without explanations, and rewards based only on final answers can introduce noise.

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

🎓University of Massachusetts AmherstAmherst Center · United States106 works · 66,692 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 worksInformation Systems · 1,437 worksComputational Theory and Mathematics · 252 works

🧭 Topics 2Research subjects and languages the work is about.Reasoning · 1,426 worksInformation retrieval & search · 473 works

Cite

Cite as Debrup Das, Sam O'Nuallain and Razieh Rahimi. 2025. RaDeR: Reasoning-aware Dense Retrieval Models. Proceedings of the 2025 Conference on Empirical Methods in Natural Language Processing.

The paper itself, not this page. BibTeX from the library's record of the work. DOI 10.18653/v1/2025.emnlp-main.1011

@inproceedings{das2025rader,
  title = {{RaDeR: Reasoning-aware Dense Retrieval Models}},
  author = {Debrup Das, Sam O'Nuallain and Razieh Rahimi},
  booktitle = {Proceedings of the 2025 Conference on Empirical Methods in Natural Language Processing},
  year = {2025},
  doi = {10.18653/v1/2025.emnlp-main.1011},
  url = {https://doi.org/10.18653/v1/2025.emnlp-main.1011}
}

References

60 references, 37 of them with a card of ours. ↗ leads to the work, «find» searches for one we have no id for yet.

  1. Akiko Aizawa, Michael Kohlhase, and Iadh Ounis. 2013. Ntcir-10 math pilot task overview. In NTCIR Conference on Evaluation of Information Access Technologies. find
  2. Akari Asai, Zeqiu Wu, Yizhong Wang, Avirup Sil, and Hannaneh Hajishirzi. 2023. Self-rag: Learning to retrieve, generate, and critique through self-reflection. Preprint, arXiv:2310.11511.
  3. Akari Asai, Zexuan Zhong, Danqi Chen, Pang Wei Koh, Luke Zettlemoyer, Hannaneh Hajishirzi, and Wen tau Yih. 2024. Reliable, adaptable, and attributable language models with retrieval. CoRR, abs/2403.03187.
  4. Payal Bajaj, Daniel Campos, Nick Craswell, Li Deng, Jianfeng Gao, Xiaodong Liu, Rangan Majumder, Andrew McNamara, Bhaskar Mitra, Tri Nguyen, Mir Rosenberg, Xia Song, Alina Stoica, Saurabh Tiwary, and Tong Wang. 2018. Ms marco: A human generated machine reading comprehension dataset. Preprint, arXiv:…
  5. Luiz Bonifacio, Hugo Abonizio, Marzieh Fadaee, and Rodrigo Nogueira. 2022. Inpars: Data augmentation for information retrieval using large language models. Preprint, arXiv:2202.05144. find
  6. Nick Craswell, Bhaskar Mitra, Emine Yilmaz, and Daniel Campos. 2021. Overview of the trec 2020 deep learning track. corr abs/2102.07662 (2021). TREC. — [arXiv:2102.07662](https://arxiv.org/abs/2102.07662) find
  7. Nick Craswell, Bhaskar Mitra, Emine Yilmaz, Daniel Campos, and Ellen M Voorhees. 2020. Overview of the trec 2019 deep learning track. TREC.
  8. Zhuyun Dai, Vincent Y Zhao, Ji Ma, Yi Luan, Jianmo Ni, Jing Lu, Anton Bakalov, Kelvin Guu, Keith Hall, and Ming-Wei Chang. 2023. Promptagator: Few-shot dense retrieval from 8 examples. In The Eleventh International Conference on Learning Representations.
  9. Kenneth Enevoldsen, Isaac Chung, Imene Kerboua, Márton Kardos, Ashwin Mathur, David Stap, Jay Gala, Wissam Siblini, Dominik Krzemiński, Genta Indra Winata, Saba Sturua, Saiteja Utpala, Mathieu Ciancone, Marion Schaeffer, Diganta Misra, Shreeya Dhakal, Jonathan Rystrøm, Roman Solomatin, Ömer Veysel… — MMTEB: Massive Multilingual Text Embedding Benchmark
  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. Aaron Grattafiori, Abhimanyu Dubey, Abhinav Jauhri, Abhinav Pandey, Abhishek Kadian, Ahmad Al-Dahle, Aiesha Letman, Akhil Mathur, Alan Schelten, Alex Vaughan, et al. 2024. The llama 3 herd of models. arXiv preprint arXiv:2407.21783.
  12. Xinyu Guan, Li Lyna Zhang, Yifei Liu, Ning Shang, Youran Sun, Yi Zhu, Fan Yang, and Mao Yang. 2025. rstar-math: Small llms can master math reasoning with self-evolved deep thinking. Preprint, arXiv:2501.04519.
  13. Kelvin Guu, Kenton Lee, Zora Tung, Panupong Pasupat, and Ming-Wei Chang. 2020. Realm: Retrieval-augmented language model pre-training. Preprint, arXiv:2002.08909.
  14. Shibo Hao, Yi Gu, Haodi Ma, Joshua Jiahua Hong, Zhen Wang, Daisy Zhe Wang, and Zhiting Hu. 2023. Reasoning with language model is planning with world model. In The 2023 Conference on Empirical Methods in Natural Language Processing. find
  15. Dan Hendrycks, Collin Burns, Saurav Kadavath, Akul Arora, Steven Basart, Eric Tang, Dawn Song, and Jacob Steinhardt. 2021. Measuring mathematical problem solving with the math dataset. Preprint, arXiv:2103.03874.
  16. Minda Hu, Licheng Zong, Hongru Wang, Jingyan Zhou, Jingjing Li, Yichen Gao, Kam-Fai Wong, Yu Li, and Irwin King. 2024. Serts: Self-rewarding tree search for biomedical retrieval-augmented generation. Preprint, arXiv:2406.11258. find
  17. Yunhai Hu, Yilun Zhao, Chen Zhao, and Arman Cohan. 2025. Mcts-rag: Enhancing retrieval-augmented generation with monte carlo tree search. Preprint, arXiv:2503.20757. find
  18. Gautier Izacard, Mathilde Caron, Lucas Hosseini, Sebastian Riedel, Piotr Bojanowski, Armand Joulin, and Edouard Grave. 2022. Unsupervised dense information retrieval with contrastive learning. Transactions on Machine Learning Research.
  19. Bowen Jin, Hansi Zeng, Zhenrui Yue, Jinsung Yoon, Sercan Arik, Dong Wang, Hamed Zamani, and Jiawei Han. 2025. Search-r1: Training llms to reason and leverage search engines with reinforcement learning. Preprint, arXiv:2503.09516.
  20. Vladimir Karpukhin, Barlas Oguz, Sewon Min, Patrick Lewis, Ledell Wu, Sergey Edunov, Danqi Chen, and Wen-tau Yih. 2020. Dense passage retrieval for open-domain question answering. In Proceedings of the 2020 Conference on Empirical Methods in Natural Language Processing (EMNLP), pages 6769–6781, Onli…
  21. Levente Kocsis and Csaba Szepesvári. 2006. Bandit based monte-carlo planning. In Machine Learning: ECML 2006, pages 282–293, Berlin, Heidelberg. Springer Berlin Heidelberg. find
  22. Jinhyuk Lee, Zhuyun Dai, Xiaoqi Ren, Blair Chen, Daniel Cer, Jeremy R. Cole, Kai Hui, Michael Boratko, Rajvi Kapadia, Wen Ding, Yi Luan, Sai Meher Karthik Duddu, Gustavo Hernandez Abrego, Weiqiang Shi, Nithi Gupta, Aditya Kusupati, Prateek Jain, Siddhartha Reddy Jonnalagadda, Ming-Wei Chang, and Ift… — Gecko: Versatile Text Embeddings Distilled from Large Language Models
  23. Yibin Lei, Liang Ding, Yu Cao, Changtong Zan, Andrew Yates, and Dacheng Tao. 2023. Unsupervised dense retrieval with relevance-aware contrastive pre-training. In Findings of the Association for Computational Linguistics: ACL 2023, pages 10932–10940, Toronto, Canada. Association for Computational Lin… find
  24. 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. 2021. Retrieval-augmented generation for knowledge-intensive nlp tasks. Preprint, arXiv:2005.11401.
  25. Jia Li, Edward Beeching, Lewis Tunstall, Ben Lipkin, Roman Soletskyi, Shengyi Huang, Kashif Rasul, Longhui Yu, Albert Q Jiang, Ziju Shen, et al. 2024. Numinamath: The largest public dataset in ai4maths with 860k pairs of competition math problems and solutions. find
  26. Zehan Li, Xin Zhang, Yanzhao Zhang, Dingkun Long, Pengjun Xie, and Meishan Zhang. 2023a. Towards general text embeddings with multi-stage contrastive learning. Preprint, arXiv:2308.03281.
  27. Zehan Li, Xin Zhang, Yanzhao Zhang, Dingkun Long, Pengjun Xie, and Meishan Zhang. 2023b. Towards general text embeddings with multi-stage contrastive learning. arXiv preprint arXiv:2308.03281.
  28. Xueguang Ma, Luyu Gao, Shengyao Zhuang, Jiaqi Samantha Zhan, Jamie Callan, and Jimmy Lin. 2025. Tevatron 2.0: Unified document retrieval toolkit across scale, language, and modality. arXiv preprint arXiv:2505.02466. — [arXiv:2505.02466](https://arxiv.org/abs/2505.02466) find
  29. Xueguang Ma, Liang Wang, Nan Yang, Furu Wei, and Jimmy Lin. 2023. Fine-tuning llama for multi-stage text retrieval. Preprint, arXiv:2310.08319.
  30. Behrooz Mansouri, Vít Novotný, Anurag Agarwal, Douglas W. Oard, and Richard Zanibbi. 2022. Overview of arqmath-3 (2022): Third clef lab on answer retrieval for questions on math (working notes version). In Conference and Labs of the Evaluation Forum. find
  31. Niklas Muennighoff, Hongjin Su, Liang Wang, Nan Yang, Furu Wei, Tao Yu, Amanpreet Singh, and Douwe Kiela. 2024. Generative representational instruction tuning. Preprint, arXiv:2402.09906.
  32. Reiichiro Nakano, Jacob Hilton, Suchir Balaji, Jeff Wu, Long Ouyang, Christina Kim, Christopher Hesse, Shantanu Jain, Vineet Kosaraju, William Saunders, Xu Jiang, Karl Cobbe, Tyna Eloundou, Gretchen Krueger, Kevin Button, Matthew Knight, Benjamin Chess, and John Schulman. 2022. Webgpt: Browser-assis…
  33. Tri Nguyen, Mir Rosenberg, Xia Song, Jianfeng Gao, Saurabh Tiwary, Rangan Majumder, and Li Deng. 2017. MS MARCO: A human-generated MAchine reading COmprehension dataset.
  34. 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
  35. Zhenting Qi, Mingyuan MA, Jiahang Xu, Li Lyna Zhang, Fan Yang, and Mao Yang. 2025. Mutual reasoning makes smaller LLMs stronger problem-solver. In The Thirteenth International Conference on Learning Representations. find
  36. Stephen Robertson, S. Walker, S. Jones, M. M. Hancock-Beaulieu, and M. Gatford. 1995. Okapi at trec-3. In Overview of the Third Text REtrieval Conference (TREC-3), pages 109–126. Gaithersburg, MD: NIST.
  37. Ohad Rubin, Jonathan Herzig, and Jonathan Berant. 2022. Learning to retrieve prompts for in-context learning. In Proceedings of the 2022 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, pages 2655–2671, Seattle, United States. As…
  38. Chris Samarinas and Hamed Zamani. 2025. Distillation and refinement of reasoning in small language models for document re-ranking. Preprint, arXiv:2504.03947. — [arXiv:2504.03947](https://arxiv.org/abs/2504.03947) find
  39. Timo Schick, Jane Dwivedi-Yu, Roberto Dessi, Roberta Raileanu, Maria Lomeli, Eric Hambro, Luke Zettlemoyer, Nicola Cancedda, and Thomas Scialom. 2023. Toolformer: Language models can teach themselves to use tools. In Thirty-seventh Conference on Neural Information Processing Systems.
  40. Rulin Shao, Rui Qiao, Varsha Kishore, Niklas Muennighoff, Xi Victoria Lin, Daniela Rus, Bryan Kian Hsiang Low, Sewon Min, Wen-tau Yih, Pang Wei Koh, et al. 2025. Reasonir: Training retrievers for reasoning tasks. arXiv preprint arXiv:2504.20595. find
  41. 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 2023, pages 9248–9274, Singapore. Association for…
  42. Weijia Shi, Sewon Min, Michihiro Yasunaga, Minjoon Seo, Richard James, Mike Lewis, Luke Zettlemoyer, and Wen-tau Yih. 2024. REPLUG: Retrieval-augmented black-box language models. In Proceedings of the 2024 Conference of the North American Chapter of the Association for Computational Linguistics: Hum…
  43. Huatong Song, Jinhao Jiang, Yingqian Min, Jie Chen, Zhipeng Chen, Wayne Xin Zhao, Lei Fang, and Ji-Rong Wen. 2025. R1-searcher: Incentivizing the search capability in llms via reinforcement learning. Preprint, arXiv:2503.05592.
  44. Hongjin Su, Weijia Shi, Jungo Kasai, Yizhong Wang, Yushi Hu, Mari Ostendorf, Wen-tau Yih, Noah A. Smith, Luke Zettlemoyer, and Tao Yu. 2023. One embedder, any task: Instruction-finetuned text embeddings. In Findings of the Association for Computational Linguistics: ACL 2023, pages 1102–1121, Toronto…
  45. Hongjin Su, Howard Yen, Mengzhou Xia, Weijia Shi, Niklas Muennighoff, Han yu Wang, Haisu Liu, Quan Shi, Zachary S. Siegel, Michael Tang, Ruoxi Sun, Jinsung Yoon, Sercan O. Arik, Danqi Chen, and Tao Yu. 2024. Bright: A realistic and challenging benchmark for reasoning-intensive retrieval. Preprint, a…
  46. Hieu Tran, Zonghai Yao, Junda Wang, Yifan Zhang, Zhichao Yang, and Hong Yu. 2024. Rare: Retrieval-augmented reasoning enhancement for large language models. Preprint, arXiv:2412.02830.
  47. Harsh Trivedi, Niranjan Balasubramanian, Tushar Khot, and Ashish Sabharwal. 2023. Interleaving retrieval with chain-of-thought reasoning for knowledge-intensive multi-step questions. Preprint, arXiv:2212.10509.
  48. Voyage. Voyage embedding models. https://docs.voyageai.com/docs/embeddings. ↗
  49. Liang Wang, Nan Yang, Xiaolong Huang, Binxing Jiao, Linjun Yang, Daxin Jiang, Rangan Majumder, and Furu Wei. 2023a. Simlm: Pre-training with representation bottleneck for dense passage retrieval. Preprint, arXiv:2207.02578.
  50. Liang Wang, Nan Yang, Xiaolong Huang, Linjun Yang, Rangan Majumder, and Furu Wei. 2024. Improving text embeddings with large language models. In Proceedings of the 62nd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers), pages 11897–11916, Bangkok, Thailand. Asso…
  51. Xuezhi Wang, Jason Wei, Dale Schuurmans, Quoc V Le, Ed H. Chi, Sharan Narang, Aakanksha Chowdhery, and Denny Zhou. 2023b. Self-consistency improves chain of thought reasoning in language models. In The Eleventh International Conference on Learning Representations.
  52. Orion Weller, Kathryn Ricci, Eugene Yang, Andrew Yates, Dawn Lawrie, and Benjamin Van Durme. 2025. Rank1: Test-time compute for reranking in information retrieval. Preprint, arXiv:2502.18418. — [arXiv:2502.18418](https://arxiv.org/abs/2502.18418) find
  53. Yuan Xia, Jingbo Zhou, Zhenhui Shi, Jun Chen, and Haifeng Huang. 2025. Improving retrieval augmented language model with self-reasoning. Proceedings of the AAAI Conference on Artificial Intelligence, 39(24):25534–25542.
  54. Chenghao Xiao, G Thomas Hudson, and Noura Al Moubayed. 2024. Rar-b: Reasoning as retrieval benchmark. Preprint, arXiv:2404.06347. — [arXiv:2404.06347](https://arxiv.org/abs/2404.06347) find
  55. An Yang, Baosong Yang, Beichen Zhang, Binyuan Hui, Bo Zheng, Bowen Yu, Chengyuan Li, Dayiheng Liu, Fei Huang, Haoran Wei, et al. 2024. Qwen2. 5 technical report. arXiv preprint arXiv:2412.15115. find
  56. Shunyu Yao, Dian Yu, Jeffrey Zhao, Izhak Shafran, Thomas L. Griffiths, Yuan Cao, and Karthik R Narasimhan. 2023a. Tree of thoughts: Deliberate problem solving with large language models. In Thirty-seventh Conference on Neural Information Processing Systems. find
  57. Shunyu Yao, Jeffrey Zhao, Dian Yu, Nan Du, Izhak Shafran, Karthik Narasimhan, and Yuan Cao. 2023b. ReAct: Synergizing reasoning and acting in language models. In International Conference on Learning Representations (ICLR).
  58. Wei Zhong, Jheng-Hong Yang, Yuqing Xie, and Jimmy Lin. 2022. Evaluating token-level and passage-level dense retrieval models for math information retrieval. In Findings of the Association for Computational Linguistics: EMNLP 2022, pages 1092–1102, Abu Dhabi, United Arab Emirates. Association for Com… find
  59. Wei Zhong, Xinyu Zhang, Ji Xin, Richard Zanibbi, and Jimmy Lin. 2021. Approach zero and anserini at the clef-2021 arqmath track: Applying substructure search and bm25 on operator tree path tokens. find
  60. Proc. CLEF 2021 (CEUR Working Notes). find

Updated

↑ back to top

This page in Markdown → https://md.research.elamur.ai/free/en/papers/das-2025-rader-reasoning-aware-dense-retrieval/