The DOI matches its Crossref record; 46 of the 46 references it cites were found in a registry or an index.
Efficiently Teaching an Effective Dense Retriever with Balanced Topic Aware Sampling
Open accessAbstract
A vital step towards the widespread adoption of neural retrieval models is their resource efficiency throughout the training, indexing and query workflows. The neural IR community made great advancements in training effective dual-encoder dense retrieval (DR) models recently. A dense text retrieval model uses a single vector representation per query and passage to score a match, which enables low-latency first-stage retrieval with a nearest neighbor search. Increasingly common, training approaches require enormous compute power, as they either conduct negative passage sampling out of a continuously updating refreshing index or require very large batch sizes. Instead of relying on more compute capability, we introduce an efficient topic-aware query and balanced margin sampling technique, called TAS-Balanced. We cluster queries once before training and sample queries out of a cluster per batch. We train our lightweight 6-layer DR model with a novel dual-teacher supervision that combines pairwise and in-batch negative teachers. Our method is trainable on a single consumer-grade GPU in under 48 hours. We show that our TAS-Balanced training method achieves state-of-the-art low-latency (64ms per query) results on two TREC Deep Learning Track query sets. Evaluated on [email protected], we outperform BM25 by 44%, a plainly trained DR by 19%, docT5query by 11%, and the previous best DR model by 5%. Additionally, TAS-Balanced produces the first dense retriever that outperforms every other method on recall at any cutoff on TREC-DL and allows more resource intensive re-ranking models to operate on fewer passages to improve results further.
Overview
The claims and statistics of this paper haven’t been checked yet.
Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
The claims and statistics of this paper haven’t been checked yet.
Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.
Self-check
Percentages, sums and reported differences re-computed from the paper’s own numbers.
The self-check needs the paper’s full text.
Citations
Cites
Vaswani, A., Shazeer, N., Parmar, N., Uszkoreit, J., Jones, L., Gomez, A. N., Kaiser, L., & Polosukhin, I. (2017) · arXiv · doi:10.48550/arxiv.1706.03762 · Cited by 120,404
Adam: A Method for Stochastic Optimization
Kingma, D. P., & Ba, J. (2014) · arXiv · doi:10.48550/arxiv.1412.6980 · Cited by 98,946
BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding
Devlin, J., Chang, M.-W., Lee, K., & Toutanova, K. (2018) · arXiv · doi:10.48550/arxiv.1810.04805 · Cited by 73,173
BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding
Devlin, J., Chang, M.-W., Lee, K., & Toutanova, K. (2019) · Proceedings of the 2019 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, Volume 1 (Long and Short Papers) · doi:10.18653/v1/n19-1423 · Cited by 54,325
(1967) · Cited by 23,719
(2017) · Cited by 11,625
Transformers: State-of-the-Art Natural Language Processing
Wolf, T., Debut, L., Sanh, V., Chaumond, J., Delangue, C., Moi, A., Cistac, P., Rault, T., Louf, R., Funtowicz, M., et al. (2020) · Proceedings of the 2020 Conference on Empirical Methods in Natural Language Processing: System Demonstrations · doi:10.18653/v1/2020.emnlp-demos.6 · Cited by 8,741
DistilBERT, a distilled version of BERT: smaller, faster, cheaper and lighter
Sanh, V., Debut, L., Chaumond, J., & Wolf, T. (2019) · arXiv · doi:10.48550/arxiv.1910.01108 · Cited by 4,653
Billion-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
Billion-scale similarity search with GPUs
Johnson, J., Douze, M., & Jégou, H. (2017) · arXiv · doi:10.48550/arxiv.1702.08734 · Cited by 4,140
Dense passage retrieval for open-domain question answering
Karpukhin, V., Oguz, B., Min, S., Lewis, P., Wu, L., Edunov, S., Chen, D., & Yih, W.-T. (2020) · Proceedings of the 2020 Conference on Empirical Methods in Natural Language Processing (EMNLP) · doi:10.18653/v1/2020.emnlp-main.550 · Cited by 3,575
HuggingFace's Transformers: State-of-the-art Natural Language Processing
Wolf, T., Debut, L., Sanh, V., Chaumond, J., Delangue, C., Moi, A., Cistac, P., Rault, T., Louf, R., Funtowicz, M., et al. (2019) · arXiv · doi:10.48550/arxiv.1910.03771 · Cited by 3,347
Unsupervised Learning of Visual Features by Contrasting Cluster Assignments
Caron, M., Misra, I., Mairal, J., Goyal, P., Bojanowski, P., & Joulin, A. (2020) · arXiv · doi:10.48550/arxiv.2006.09882 · Cited by 2,499
MS MARCO: A Human Generated MAchine Reading COmprehension Dataset
Bajaj, P., Campos, D., Craswell, N., Deng, L., Gao, J., Liu, X., Majumder, R., McNamara, A., Mitra, B., Nguyen, T., et al. (2016) · arXiv · doi:10.48550/arxiv.1611.09268 · Cited by 2,473
Learning to rank: from pairwise approach to listwise approach
Cao, Z., Qin, T., Liu, T.-Y., Tsai, M.-F., & Li, H. (2007) · Proceedings of the 24th international conference on Machine learning · doi:10.1145/1273496.1273513 · Cited by 1,960 · cited 1×
“We study two different types of loss functions: First are list-based losses (KL Divergence, ListNet [2] and, the LambdaLoss version nDCG2 [39]) where we build a ranked list of in-batch and pairwise negatives per query and second the pairwise Margin-MSE loss that repeats the relevant passage per query to pair with all…”
MethodTinyBERT: Distilling BERT for natural language understanding
Jiao, X., Yin, Y., Shang, L., Jiang, X., Chen, X., Li, L., Wang, F., & Liu, Q. (2020) · Findings of the Association for Computational Linguistics: EMNLP 2020 · doi:10.18653/v1/2020.findings-emnlp.372 · Cited by 1,786
ColBERT: Efficient and Effective Passage Search via Contextualized Late Interaction over BERT
Khattab, O., & Zaharia, M. (2020) · Proceedings of the 43rd International ACM SIGIR Conference on Research and Development in Information Retrieval · doi:10.1145/3397271.3401075 · Cited by 1,301
Transformers: State-of-the-Art Natural Language Processing
Wolf, T., Debut, L., Sanh, V., D., M., Châtel, G., Chaumond, J., Rault, T., Voss, C., Wang, F., et al. (2019) · Zenodo (CERN European Organization for Nuclear Research) · doi:10.5281/zenodo.3385997 · Cited by 1,037
Nogueira, R., & Cho, K. (2019) · arXiv · doi:10.48550/arxiv.1901.04085 · Cited by 615
The use of hierarchic clustering in information retrieval
Jardine, N., & van Rijsbergen, C. J. (1971) · Information Processing & Management · doi:10.1016/0020-0271(71)90051-9 · Cited by 603
How reliable are the results of large-scale information retrieval experiments?
Zobel, J. (1998) · Proceedings of the 21st annual international ACM SIGIR conference on Research and development in information retrieval · doi:10.1145/290941.291014 · Cited by 579 · cited 1×
“Naturally, most queries have fewer relevant passages than non-relevant ones [45].”
BackgroundSparse, Dense, and Attentional Representations for Text Retrieval
Luan, Y., Eisenstein, J., Toutanova, K., & Collins, M. (2021) · Transactions of the Association for Computational Linguistics · doi:10.1162/tacl_a_00369 · Cited by 363 · cited 1×
“2 BERTDOT Dense Retrieval Model The BERTDOT model encodes query q1:m and passage p1:n sequences independently from each other and matches only a single representation vector of the query with a single representation vector of a passage [24, 25, 41].”
BackgroundAnserini: Enabling the Use of Lucene for Information Retrieval Research
Yang, P., Fang, H., & Lin, J. (2017) · Proceedings of the 40th International ACM SIGIR Conference on Research and Development in Information Retrieval · doi:10.1145/3077136.3080721 · Cited by 319
Self-citationContext-Aware Sentence/Passage Term Importance Estimation For First Stage RetrievalPreprint
Dai, Z., & Callan, J. (2019) · arXiv (Cornell University) · doi:10.48550/arxiv.1910.10687 · Cited by 258
Hard Negative Mixing for Contrastive Learning
Kalantidis, Y., Sarıyıldız, M. B., Pion, N., Weinzaepfel, P., & Larlus, D. (2020) · arXiv (Cornell University) · doi:10.48550/arxiv.2010.01028 · Cited by 257
CEDR: Contextualized Embeddings for Document Ranking
MacAvaney, S., Yates, A., Cohan, A., & Goharian, N. (2019) · Proceedings of the 42nd International ACM SIGIR Conference on Research and Development in Information Retrieval · doi:10.1145/3331184.3331317 · Cited by 239
The Expando-Mono-Duo Design Pattern for Text Ranking with Pretrained Sequence-to-Sequence ModelsPreprint
Pradeep, R., Nogueira, R. F., & Lin, J. (2021) · arXiv (Cornell University) · doi:10.48550/arxiv.2101.05667 · Cited by 237
Self-citationApproximate Nearest Neighbor Negative Contrastive Learning for Dense Text Retrieval
Xiong, L., Xiong, C., Li, Y., Tang, K.-F., Liu, J., Bennett, P., Ahmed, J., & Overwijk, A. (2020) · arXiv · doi:10.48550/arxiv.2007.00808 · Cited by 204
Rethink Training of BERT Rerankers in Multi-stage Retrieval Pipeline
Gao, L., Dai, Z., & Callan, J. (2021) · Lecture notes in computer science · doi:10.1007/978-3-030-72240-1_26 · Cited by 177
Ranking Distillation: Learning Compact Ranking Models With High Performance for Recommender System
Tang, J., & Wang, K. (2018) · Proceedings of the 24th ACM SIGKDD International Conference on Knowledge Discovery & Data Mining · doi:10.1145/3219819.3220021 · Cited by 167
Distilling Dense Representations for Ranking using Tightly-Coupled TeachersPreprint
Lin, S.-C., Yang, J.-H., & Lin, J. (2020) · arXiv (Cornell University) · doi:10.48550/arxiv.2010.11386 · Cited by 150
Self-citationCross-Domain Modeling of Sentence-Level Evidence for Document Retrieval
Yilmaz, Z. A., Yang, W., Zhang, H., & Lin, J. (2019) · Conference on Empirical Methods in Natural Language Processing (EMNLP) · doi:10.18653/v1/d19-1352 · Cited by 144
Self-citationThe LambdaLoss Framework for Ranking Metric Optimization
Wang, X., Li, C., Golbandi, N., Bendersky, M., & Najork, M. (2018) · Proceedings of the 27th ACM International Conference on Information and Knowledge Management · doi:10.1145/3269206.3271784 · Cited by 136
Overview of the TREC 2020 Deep Learning Track
Craswell, N., Mitra, B., Yilmaz, E., & Campos, D. (2021) · arXiv · doi:10.48550/arxiv.2102.07662 · Cited by 107
Distilling Knowledge from Reader to Retriever for Question Answering
Izacard, G., & Grave, E. (2020) · arXiv · doi:10.48550/arxiv.2012.04584 · Cited by 102 · cited 2×
“Apart from specifically training dense retrieval models, knowledge distillation gains popularity, with general-purpose BERT-style models [17, 33] as well as a range of applications in IR: from sequential recommendation models [35], BERT-based retrieval chatbots [36], BERT-based Question Answering [15], reducing the si…”
BackgroundMethodOverview of the TREC 2021 deep learning trackPreprint
Craswell, N., Mitra, B., Yılmaz, E., Campos, D., & Lin, J. (2025) · arXiv (Cornell University) · doi:10.48550/arxiv.2507.08191 · Cited by 69
Self-citationOverview of the TREC 2019 deep learning track
Craswell, N., Mitra, B., Yilmaz, E., Campos, D., & Voorhees, E. M. (2020) · arXiv · doi:10.48550/arxiv.2003.07820 · Cited by 68
Qu, Y., Ding, Y., Liu, J., Liu, K., Ren, R., Zhao, W. X., Dong, D., Wu, H. R., & Wang, H. (2020) · arXiv (Cornell University) · doi:10.48550/arxiv.2010.08191 · Cited by 66
Improving Efficient Neural Ranking Models with Cross-Architecture Knowledge Distillation
Hofstätter, S., Althammer, S., Schröder, M., Sertkan, M., & Hanbury, A. (2020) · Zenodo (CERN European Organization for Nuclear Research) · doi:10.5281/zenodo.4068215 · Cited by 54
Self-citationTwinBERT: Distilling Knowledge to Twin-Structured BERT Models for Efficient RetrievalPreprint
Lu, W., Jiao, J., & Zhang, R. (2020) · arXiv (Cornell University) · doi:10.48550/arxiv.2002.06275 · Cited by 52
Learning To Retrieve: How to Train a Dense Retrieval Model Effectively and EfficientlyPreprint
Zhan, J., Mao, J., Liu, Y., Zhang, M., & Ma, S. (2020) · arXiv (Cornell University) · doi:10.48550/arxiv.2010.10469 · Cited by 50
Simplified TinyBERT: Knowledge Distillation for Document Retrieval
Chen, X., He, B., Hui, K., Sun, L., & Sun, Y. (2021) · Lecture notes in computer science · doi:10.1007/978-3-030-72240-1_21 · Cited by 46
Distilling Knowledge for Fast Retrieval-based Chat-bots
Tahami, A. V., Ghajar, K., & Shakery, A. (2020) · International ACM SIGIR Conference on Research and Development in Information Retrieval (SIGIR) · doi:10.1145/3397271.3401296 · Cited by 32
Rethinking Image Mixture for Unsupervised Visual Representation LearningPreprint
Shen, Z., Liu, Z., Liu, Z., Savvides, M., & Darrell, T. (2020) · arXiv (Cornell University) · Cited by 25
Training Curricula for Open Domain Answer Re-Ranking
MacAvaney, S., Nardini, F. M., Perego, R., Tonellotto, N., Goharian, N., & Frieder, O. (2020) · International ACM SIGIR Conference on Research and Development in Information Retrieval (SIGIR) · doi:10.1145/3397271.3401094 · Cited by 19
Learning a Better Negative Sampling Policy with Deep Neural Networks for Search
Cohen, D., Jordan, S. M., & Croft, W. B. (2019) · ACM SIGIR International Conference on Theory of Information Retrieval (ICTIR) · doi:10.1145/3341981.3344220 · Cited by 18
Cited by
Precise Zero-Shot Dense Retrieval without Relevance Labels
Gao, L., Ma, X., Lin, J., & Callan, J. (2023) · Proceedings of the 61st Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers) · doi:10.18653/v1/2023.acl-long.99 · Cited by 362
Self-citationPrecise Zero-Shot Dense Retrieval without Relevance Labels
Gao, L., Ma, X., Lin, J., & Callan, J. (2022) · arXiv · doi:10.48550/arxiv.2212.10496 · Cited by 362
Self-citationColBERTv2: Effective and Efficient Retrieval via Lightweight Late Interaction
Santhanam, K., Khattab, O., Saad-Falcon, J., Potts, C., & Zaharia, M. (2022) · Proceedings of the 2022 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies · doi:10.18653/v1/2022.naacl-main.272 · Cited by 353
Dense Text Retrieval Based on Pretrained Language Models: A Survey
Zhao, W. X., Liu, J., Ren, R., & Wen, J.-R. (2024) · ACM Transactions on Information Systems · doi:10.1145/3637870 · Cited by 211
Zhao, S., Yang, Y., Wang, Z., He, Z., Qiu, L. K., & Qiu, L. (2024) · arXiv · doi:10.48550/arxiv.2409.14924 · Cited by 180
Metadata
- Authors
- Hofstätter, S., Lin, S.-C., Yang, J.-H., Lin, J., & Hanbury, A.
- Venue
- Proceedings of the 44th International ACM SIGIR Conference on Research and Development in Information Retrieval
- Publisher
- ACM
- Published
- 2021-07-11
- Pages
- 113-122
- OpenAlex
- W3153015371
- Cited by
- 293
- References
- 46
- FWCI
- 46.14
- Type
- conference paper
- Language
- en
- Open access
- green
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
Data sources
Bibliographic data: Semantic Scholar, OpenAlex, Crossref.
Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).
Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.