In-Batch Negatives for Knowledge Distillation with Tightly-Coupled Teachers for Dense Retrieval
Open accessThe DOI matches its Crossref record; 15 of the 15 references it cites were found in a registry or an index.
Abstract
We present an efficient training approach to text retrieval with dense representations that applies knowledge distillation using the Col-BERT late-interaction ranking model.Specifically, we propose to transfer the knowledge from a bi-encoder teacher to a student by distilling knowledge from ColBERT's expressive MaxSim operator into a simple dot product.The advantage of the bi-encoder teacherstudent setup is that we can efficiently add inbatch negatives during knowledge distillation, enabling richer interactions between teacher and student models.In addition, using Col-BERT as the teacher reduces training cost compared to a full cross-encoder.Experiments on the MS MARCO passage and document ranking tasks and data from the TREC 2019 Deep Learning Track demonstrate that our approach helps models learn robust representations for dense retrieval effectively and efficiently.
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
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
Sentence-BERT: Sentence Embeddings using Siamese BERT-Networks
Reimers, N., & Gurevych, I. (2019) · Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing and the 9th International Joint Conference on Natural Language Processing (EMNLP-IJCNLP) · doi:10.18653/v1/d19-1410 · Cited by 12,778
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 · cited 1×
“To characterize end-to-end effectiveness and efficiency, we perform sparse retrieval with the Py-serini toolkit (Lin et al., 2021) and dense retrieval with Faiss (Johnson et al., 2017), but implement the score combination in separate custom code.”
MethodDense 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
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
Malkov, Y. A., & Yashunin, D. A. (2020) · IEEE Transactions on Pattern Analysis and Machine Intelligence · doi:10.1109/tpami.2018.2889473 · Cited by 1,904
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 · cited 5×
“For example, ColBERT (Khattab and Zaharia, 2020) requires storing all the WordPiece token vectors of each text (passage) in the corpus.”
BackgroundMethodInfluentialLatent retrieval for weakly supervised open domain question answering
Lee, K., Chang, M.-W., & Toutanova, K. (2019) · Proceedings of the 57th Annual Meeting of the Association for Computational Linguistics · doi:10.18653/v1/p19-1612 · Cited by 771 · cited 2×
“The bi-encoder design has been widely adopted for dense retrieval (Lee et al., 2019; Chang et al., 2020; Guu et al., 2020; Karpukhin et al., 2020; Luan et al., 2021; Qu et al., 2020; Xiong et al., 2021), where queries and passages are encoded in a low-dimensional space.”
Method
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-citationOne-Shot Labeling for Automatic Relevance Estimation
MacAvaney, S., & Soldaini, L. (2023) · Proceedings of the 46th International ACM SIGIR Conference on Research and Development in Information Retrieval · doi:10.1145/3539618.3592032 · Cited by 65
“For lexical similarity, we use the top k = 128 BM25 results for the given document, and for semantic similarity, we use the top k = 128 nearest neighbors over TCT-ColBERT [17] embeddings5 — both of which have been recently shown as an effective signal to identify additional relevant documents when”
MethodPromptagator: Few-shot Dense Retrieval From 8 Examples
Dai, Z., Zhao, V. Y., Ma, J., Luan, Y., Ni, J., Lu, J., Bakalov, A., Guu, K., Hall, K. B., & Chang, M.-W. (2022) · arXiv · doi:10.48550/arxiv.2209.11755 · Cited by 43
“Distilling interaction based models into representation based models has been shown effective in closing the gap between the two (Hofstätter et al., 2020; Ren et al., 2021a; Lin et al., 2021; Ren et al., 2021b; Reddi et al., 2021; Zhang et al., 2022).”
MethodPre-train a Discriminative Text Encoder for Dense Retrieval via Contrastive Span Prediction
Ma, X., Guo, J., Zhang, R., Fan, Y., & Cheng, X. (2022) · Proceedings of the 45th International ACM SIGIR Conference on Research and Development in Information Retrieval · doi:10.1145/3477495.3531772 · Cited by 35
Metadata
- Authors
- Lin, S.-C., Yang, J.-H., & Lin, J.
- Venue
- Proceedings of the 6th Workshop on Representation Learning for NLP (RepL4NLP-2021)
- Published
- 2021
- Pages
- 163-173
- OpenAlex
- W3184918446
- Cited by
- 122
- References
- 15
- FWCI
- 21.65
- Type
- conference paper
- Language
- en
- Open access
- hybrid
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
How to cite
Cite CoChat for the checks on this page, and the paper for its own findings.
- This page
- CoChat. (2026). Citation check: In-Batch Negatives for Knowledge Distillation with Tightly-Coupled Teachers for Dense Retrieval. https://app.cochat.ai/papers/10.18653/v1/2021.repl4nlp-1.17
- The paper
- Lin, S.-C., Yang, J.-H., & Lin, J. (2021). In-Batch Negatives for Knowledge Distillation with Tightly-Coupled Teachers for Dense Retrieval. Proceedings of the 6th Workshop on Representation Learning for NLP (RepL4NLP-2021). https://doi.org/10.18653/v1/2021.repl4nlp-1.17
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.