Reciprocal rank fusion outperforms condorcet and individual rank learning methods
Abstract
Reciprocal Rank Fusion (RRF), a simple method for combining the document rankings from multiple IR systems, consistently yields better results than any individual system, and better results than the standard method Condorcet Fuse. This result is demonstrated by using RRF to combine the results of several TREC experiments, and to build a meta-learner that ranks the LETOR 3 dataset better than any previously reported method
Integrity signals
Registry
Cites
Statistics
Claims the paper makes
Claims other papers make about this one
Each is checked against this paper’s own text: does it say what the citing paper says it does?
The combination of vector and BM25 retrieval through RRF fusion consistently outperforms any single approach.
“The combination of vector and BM25 [Robertson and Zaragoza, 2009] retrieval through RRF fusion [Cormack et al., 2009] consistently outperforms any single approach, consistent with findings in hybrid search literature [Ma et al., 2021].”
From Cognis: Context-Aware Memory for Conversational AI Agents (2026) · cites this paper as Cormack et al., 2009
Cognis has a hybrid retrieval pipeline using RRF fusion (70% vector + 30% BM25), explicit temporal boosting for time-sensitive queries, and a BGE-2 cross-encoder reranker for final result refinement.
“A hybrid retrieval pipeline using RRF fusion [Cormack et al., 2009] (70% vector + 30% BM25), explicit temporal boosting for time-sensitive queries, and a BGE-2 [Chen et al., 2024] cross-encoder reranker for final result refinement”
From Cognis: Context-Aware Memory for Conversational AI Agents (2026) · cites this paper as Cormack et al., 2009
Reciprocal Rank Fusion (RRF) provides a simple, parameter-light method for combining ranked lists from heterogeneous retrievers, making it well-suited to systems that pair vector and full-text indexes.
“Reciprocal Rank Fusion (RRF) [Cormack et al., 2009] provides a simple, parameter-light method for combining ranked lists from heterogeneous retrievers, making it well-suited to systems that pair vector and full-text indexes.”
From MemX: A Local-First Long-Term Memory System for AI Assistants (2026) · cites this paper as Cormack et al., 2009
Cited by
Sorted by citations
Matching patients to clinical trials with large language models
Jin, Q., Wang, Z., Floudas, C. S., Chen, F., Gong, C., Bracken-Clarke, D., Xue, E., Yang, Y., Sun, J., & Lu, Z. (2024) · Nature Communications · doi:10.1038/s41467-024-53081-z · Cited by 287 · see citing works
Generation-Augmented Retrieval for Open-Domain Question Answering
Mao, Y., He, P., Liu, X., Shen, Y., Gao, J., Han, J., & Chen, W. (2021) · Proceedings of the 59th Annual Meeting of the Association for Computational Linguistics and the 11th International Joint Conference on Natural Language Processing (Volume 1: Long Papers) · doi:10.18653/v1/2021.acl-long.316 · Cited by 146 · see citing works
Lin, S.-C., Yang, J.-H., Nogueira, R., Tsai, M.-F., Wang, C.-J., & Lin, J. (2020) · arXiv · doi:10.48550/arxiv.2005.02230 · Cited by 62 · see citing works
Generation-Augmented Retrieval for Open-domain Question Answering
Mao, Y., He, P., Liu, X., Shen, Y., Gao, J., Han, J., & Chen, W. (2020) · arXiv · doi:10.48550/arxiv.2009.08553 · Cited by 31 · see citing works
A Comparative Analysis of Decision-Level Fusion for Multimodal Driver Behaviour Understanding
Roitberg, A., Peng, K., Marinov, Z., Seibold, C., Schneider, D., & Stiefelhagen, R. (2022) · 2022 IEEE Intelligent Vehicles Symposium (IV) · doi:10.1109/iv51971.2022.9827426 · Cited by 20 · see citing works
Zep: A Temporal Knowledge Graph Architecture for Agent Memory
Rasmussen, P., Paliychuk, P., Beauvais, T., Ryan, J., & Chalef, D. (2025) · arXiv · doi:10.48550/arxiv.2501.13956 · Cited by 9 · see citing works
“Zep supports existing reranking approaches such as Reciprocal Rank Fusion (RRF) [20] and Maximal Marginal Relevance (MMR) [21].”
MeshAgent: enabling reliable network management with large language models
Zhou, Y., Hsieh, K., Mani, S. K., Kandula, S., & Liu, Z. (2025) · Proceedings of the ACM on Measurement and Analysis of Computing Systems · doi:10.1145/3771567 · Cited by 7 · see citing works
MAGMA: A Multi-Graph based Agentic Memory Architecture for AI Agents
Jiang, D., Li, Y., Li, G., & Li, B. (2026) · Proceedings of the 64th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers) · doi:10.18653/v1/2026.acl-long.1709 · Cited by 2 · see citing works
Żatuchin, D. (2026) · Research Square (preprint) · doi:10.21203/rs.3.rs-9801639/v1
Lewis, M., Thio, S., Roberts, A., Siju, C., Mukit, W., Kuruvilla, R., Jiang, Z. J., Möller-Grell, N., Borakati, A., Dobson, R. J., et al. (2025) · arXiv · doi:10.48550/arxiv.2510.02967
“To leverage the complementary strengths of both sparse and dense retrieval, a hybrid search approach is implemented using a Weighted Reciprocal Rank Fusion (Cormack et al., 2009).”
D-MEM: Dopamine-Gated Agentic Memory via Reward Prediction Error Routing
Song, Y., & Xin, Q. (2026) · arXiv · doi:10.48550/arxiv.2603.14597 · Cited by 1 · see citing works
“…work explored dense passage retrieval [Karpukhin et al., 2020] for richer semantic matching and sparse lexical retrieval via BM25 [Robertson and Zaragoza, 2009] for precise entity-level lookup; hybrid fusion of both [Cormack et al., 2009] has become the practical default in high-precision systems.”
MemX: A Local-First Long-Term Memory System for AI Assistants
Sun, L. (2026) · arXiv · doi:10.48550/arxiv.2603.16171
“Reciprocal Rank Fusion (RRF) [Cormack et al., 2009] provides a simple, parameter-light method for combining ranked lists from heterogeneous retrievers, making it well-suited to systems that pair vector and full-text indexes.”
Cognis: Context-Aware Memory for Conversational AI Agents
Daftari, P., Patel, K., Kapale, S., George, J., & Surendira, S. (2026) · arXiv · doi:10.48550/arxiv.2604.19771
“We combine results from vector and BM25 search using Reciprocal Rank Fusion (RRF) [Cormack et al., 2009]: where k = 10 is a smoothing constant and rank r ( d ) is the rank of document d in retriever r .”
Hindsight: Structured Agent Memory that Retains, Recalls, and Reflects
Latimer, C., Boschi, N., Neeser, A., Bartholomew, C., Srivastava, G., Wang, X., & Ramakrishnan, N. (2026) · Proceedings of the 64th Annual Meeting of the Association for Computational Linguistics (Volume 3: System Demonstrations) · doi:10.18653/v1/2026.acl-demo.27
“Results are merged using Reciprocal Rank Fusion (RRF) (Cormack et al., 2009) and refined with a cross-encoder reranker (Nogueira and Cho, 2019).”
InfluentialAi, Q., Zhan, J., & Liu, Y. (2025) · arXiv · doi:10.48550/arxiv.2501.02842
Singh, A., & Das, R. (2026) · Proceedings of the 20th International Workshop on Semantic Evaluation (2026) · doi:10.18653/v1/2026.semeval-1.393
- Authors
- Cormack, G. V., Clarke, C. L. A., & Buettcher, S.
- Affiliations
- University of Waterloo (CA)
- Google, Mountain View, CA, USA
- Venue
- Proceedings of the 32nd international ACM SIGIR conference on Research and development in information retrieval
- Publisher
- ACM
- Published
- 2009-07-19
- Pages
- 758-759
- OpenAlex
- W2148972377
- Cited by
- 797
- References
- 0
- Citations by year
- 2026: 297 · 2025: 92 · 2024: 57 · 2023: 54 · 2022: 30 · 2021: 40 · 2020: 41 · 2019: 29 · 2018: 25 · 2017: 27 · 2016: 23 · 2015: 22 · 2014: 20 · 2013: 15 · 2012: 14
- FWCI
- 6.21
- Type
- conference paper
- Language
- en
- Open access
- closed
- Locations
- International ACM SIGIR Conference on Research and Development in Information Retrieval (SIGIR) (publishedVersion)
- Topics
- Physical Sciences › Computer Science › Information Systems › Information Retrieval and Search Behavior
- Physical Sciences › Computer Science › Information Systems › Recommender Systems and Techniques
- Physical Sciences › Computer Science › Artificial Intelligence › Text and Document Classification Technologies
- Keywords
- reciprocal rank fusion, meta-learner, information retrieval evaluation
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
