Scalable and Effective Generative Information Retrieval
Open accessAbstract
Recent research has shown that transformer networks can be used as differentiable search indexes by representing each document as a sequence of document ID tokens. These generative retrieval models cast the retrieval problem to a document ID generation problem for each query. Despite their elegant design, existing generative retrieval models only perform well on artificially-constructed and small-scale collections. This paper represents an important milestone in generative retrieval research by showing that generative retrieval models can be trained to perform effectively on large-scale standard retrieval benchmarks. In more detail, we propose RIPOR- an optimization framework for generative retrieval that is designed based on two often-overlooked fundamental design considerations. First, RIPOR introduces a novel prefix-oriented ranking optimization algorithm for accurate estimation of relevance score during sequential document ID generation. Second, RIPOR constructs document IDs based on the relevance associations between queries and documents. Evaluation on MSMARCO and TREC Deep Learning Track reveals that RIPOR surpasses state-of-the-art generative retrieval models by a large margin (e.g., 30.5% MRR improvements on MS MARCO Dev Set).
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?
GR models with explicit ID tokens have less flexibility and capability in document modeling than dense retrieval vectors, and their retrieval performance is usually poor.
“Compared to vectors in dense retrieval, these methods have less flexibility and capability in document modeling as they discretize document semantic meanings with limited number of tokens, and their retrieval performance is usually poor [140].”
From Foundations of GenIR (2025) · cites this paper as [140]
Cites
Exploring the Limits of Transfer Learning with a Unified Text-to-Text Transformer
Raffel, C., Shazeer, N., Roberts, A., Lee, K., Narang, S., Matena, M., Zhou, Y., Li, W., & Liu, P. J. (2019) · arXiv · doi:10.48550/arxiv.1910.10683 · Cited by 10,929 · see citing works
Bengio, Y., Louradour, J., Collobert, R., & Weston, J. (2009) · Proceedings of the 26th Annual International Conference on Machine Learning · doi:10.1145/1553374.1553380 · Cited by 5,366 · cited 1× · see citing works
“To better learn representations aligned with the left-to-right decoding characteristic of the beam search, we draw inspiration from curriculum learning [2, 25, 27, 49] and implement a progressive training strategy.”
MethodBillion-scale similarity search with GPUs
Johnson, J., Douze, M., & Jegou, H. (2021) · IEEE Transactions on Big Data · doi:10.1109/tbdata.2019.2921572 · Cited by 4,228 · see citing works
Natural Questions: A benchmark for question answering research
Kwiatkowski, T., Palomaki, J., Redfield, O., Collins, M., Parikh, A., Alberti, C., Epstein, D., Polosukhin, I., Devlin, J., Lee, K., et al. (2019) · Transactions of the Association for Computational Linguistics · doi:10.1162/tacl_a_00276 · Cited by 2,069 · see citing works
Optimizing Dense Retrieval Model Training with Hard Negatives
Zhan, J., Mao, J., Liu, Y., Guo, J., Zhang, M., & Ma, S. (2021) · arXiv · doi:10.48550/arxiv.2104.08051 · Cited by 251 · see citing works
Optimizing Dense Retrieval Model Training with Hard Negatives
Zhan, J., Mao, J., Liu, Y., Guo, J., Zhang, M., & Ma, S. (2021) · Proceedings of the 44th International ACM SIGIR Conference on Research and Development in Information Retrieval · doi:10.1145/3404835.3462880 · Cited by 248 · see citing works
Continual Learning for Generative Retrieval over Dynamic Corpora
Chen, J., Zhang, R., Guo, J., de Rijke, M., Chen, W., Fan, Y., & Cheng, X. (2023) · Proceedings of the 32nd ACM International Conference on Information and Knowledge Management · doi:10.1145/3583780.3614821 · Cited by 40 · see citing works
A Unified Generative Retriever for Knowledge-Intensive Language Tasks via Prompt Learning
Chen, J., Zhang, R., Guo, J., de Rijke, M., Liu, Y., Fan, Y., & Cheng, X. (2023) · Proceedings of the 46th International ACM SIGIR Conference on Research and Development in Information Retrieval · doi:10.1145/3539618.3591631 · Cited by 32 · see citing works
Flagged first, then by citations
Cited by
Sorted by citations
Zeng, H., Luo, C., & Zamani, H. (2024) · Proceedings of the 47th International ACM SIGIR Conference on Research and Development in Information Retrieval · doi:10.1145/3626772.3657746 · Cited by 22 · see citing works
Self-citationLightweight and Direct Document Relevance Optimization for Generative Information Retrieval
Mekonnen, K. A., Tang, Y., & de Rijke, M. (2025) · Proceedings of the 48th International ACM SIGIR Conference on Research and Development in Information Retrieval · doi:10.1145/3726302.3730023 · Cited by 3 · see citing works
“The generation process is controlled through (constrained) beam search [33, 52, 56, 62, 63, 72].”
InfluentialMemory in the Age of AI Agents
Hu, Y., Liu, S., Yue, Y., Zhang, G., Liu, B., Zhu, F., Lin, J., Guo, H., Dou, S., Xi, Z., et al. (2025) · arXiv · doi:10.48550/arxiv.2512.13564 · Cited by 1 · see citing works
Ai, Q., Zhan, J., & Liu, Y. (2025) · arXiv · doi:10.48550/arxiv.2501.02842
“Compared to vectors in dense retrieval, these methods have less flexibility and capability in document modeling as they discretize document semantic meanings with limited number of tokens, and their retrieval performance is usually poor [140].”
- Authors
- Zeng, H., Luo, C., Jin, B., Sarwar, S. M., Wei, T., & Zamani, H.
- Affiliations
- University of Massachusetts Amherst (US)
- Amazon (United States) (US)
- University of Illinois Urbana-Champaign (US)
- Venue
- Proceedings of the ACM Web Conference 2024
- Publisher
- ACM
- Published
- 2024-05-13
- Pages
- 1441-1452
- OpenAlex
- W4396723267
- Cited by
- 32
- References
- 8
- Citations by year
- 2026: 9 · 2025: 17 · 2024: 6
- FWCI
- 12.18
- Type
- conference paper
- Language
- en
- Open access
- hybrid
- doi.org
- Licence
- Locations
- International World Wide Web Conference (WWW) (publishedVersion, cc-by)
- arXiv (Cornell University) (submittedVersion, cc-by)
- arXiv (Cornell University) (cc-by)
- Topics
- Physical Sciences › Computer Science › Information Systems › Information Retrieval and Search Behavior
- Physical Sciences › Computer Science › Artificial Intelligence › Topic Modeling
- Physical Sciences › Computer Science › Computer Vision and Pattern Recognition › Advanced Image and Video Retrieval Techniques
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
