ColBERT: Efficient and Effective Passage Search via Contextualized Late Interaction over BERT
The DOI matches its Crossref record; 26 of the 26 references it cites were found in a registry or an index.
Abstract
Recent progress in Natural Language Understanding (NLU) is driving fast-paced advances in Information Retrieval (IR), largely owed to fine-tuning deep language models (LMs) for document ranking. While remarkably effective, the ranking models based on these LMs increase computational cost by orders of magnitude over prior approaches, particularly as they must feed each query-document pair through a massive neural network to compute a single relevance score. To tackle this, we present ColBERT, a novel ranking model that adapts deep LMs (in particular, BERT) for efficient retrieval. ColBERT introduces a late interaction architecture that independently encodes the query and the document using BERT and then employs a cheap yet powerful interaction step that models their fine-grained similarity. By delaying and yet retaining this fine-granular interaction, ColBERT can leverage the expressiveness of deep LMs while simultaneously gaining the ability to pre-compute document representations offline, considerably speeding up query processing. Crucially, ColBERT's pruning-friendly interaction mechanism enables leveraging vector-similarity indexes for end-to-end retrieval directly from millions of documents. We extensively evaluate ColBERT using two recent passage search datasets. Results show that ColBERT's effectiveness is competitive with existing BERT-based models (and outperforms every non-BERT baseline), while executing two orders-of-magnitude faster and requiring up to four orders-of-magnitude fewer FLOPs per query.
Overview
Integrity signals
Registry
Cites
Statistics
Verification
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?
In the citing paper
“The only ML component is a CrossEncoder reranker ( mxbai-rerank-large-v1 , 435M parameters, DeBERTaV3) fused with ColBERT (Khattab & Zaharia, 2020) via Reciprocal Rank Fusion (RRF), applied after retrieval is complete to prioritize passages before token-budget truncation.”
From SmartSearch: How Ranking Beats Structure for Conversational Memory Retrieval (2026) · cites this paper as Khattab & Zaharia, 2020
In the citing paper
“While dense retrievers like Dense Passage Retrieval (DPR) [14] and late-interaction models like ColBERT [15] improve recall on single-hop queries, they often struggle with multi-hop reasoning, where the answer depends on composing information from multiple disjoint documents [20, 6].”
From Replace, Don't Expand: Mitigating Context Dilution in Multi-Hop RAG via Fixed-Budget Evidence Assembly (2025) · cites this paper as [15]
In the citing paper
“Research on retrieval optimization has explored multiple techniques, yet studies specifically focused on prefetching size in RaLM systems are still limited.”
From Exploring Optimal Prefetching Sizes in RaLMSpec to Enhance Retrieval-Augmented Generation Efficiency (2025) · cites this paper as [9]
In the citing paper
“Early RAG implementations, however, often relied on document retrieval methods that introduced noise and excessive context [26, 55, 60, 122].”
From LEGO-GraphRAG: Modularizing Graph-Based Retrieval-Augmented Generation for Design Space Exploration (2025) · cites this paper as [55]
In the citing paper
“Standard retrieval methods like RAG and MemoryBank (Zhong et al., 2024) rely fundamentally on vector similarity (Karpukhin et al., 2020; Khattab and Zaharia, 2020), representing memories as isolated points in embedding space (Hu et al., 2025). Consequently, they struggle with queries requiring causal bridging between semantically dissimilar or distant events (Yang et al., 2018; Qi et al., 2019; Tr”
From Synapse: Empowering LLM Agents with Episodic-Semantic Memory via Spreading Activation (2026) · cites this paper as Khattab and Zaharia, 2020
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 · cited 1×
“At a high level, TK improves the KNRM architecture described earlier: while KNRM employs kernel pooling on top of word-embedding-based interaction, TK uses a Transformer [34] component for contextually encoding queries and documents before kernel pooling.”
MethodBERT: 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 · cited 2×
“Among them, a recent approach has emerged that ne-tunes deep pre-trained language models (LMs) like ELMo [29] and BERT [5] for estimating relevance.”
BackgroundGoogle's Neural Machine Translation System: Bridging the Gap between Human and Machine Translation
Wu, Y., Schuster, M., Chen, Z., Le, Q. V., Norouzi, M., Macherey, W., Krikun, M., Cao, Y., Gao, Q., Macherey, K., et al. (2016) · arXiv · doi:10.48550/arxiv.1609.08144 · Cited by 5,319
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 2×
“Additionally, it enables ColBERT to leverage vector-similarity search indexes (e.g., [1, 16]) to retrieve the top-k results directly from a large document collection, substantially improving recall over models that only re-rank the output of term-based retrieval.”
MethodMS 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
Robertson, S. E., Walker, S., Jones, S., Hancock‐Beaulieu, M., & Gatford, M. (1994) · Text REtrieval Conference · Cited by 2,212
Learning deep structured semantic models for web search using clickthrough data
Huang, P.-S., He, X., Gao, J., Deng, L., Acero, A., & Heck, L. (2013) · Proceedings of the 22nd ACM international conference on Information & Knowledge Management · doi:10.1145/2505515.2505665 · Cited by 2,009 · cited 1×
“On the le, Figure 2 (a) illustrates representation-focused rankers, which independently compute an embedding for q and another for d and estimate relevance as a single similarity score between two vectors [13, 41].”
BackgroundTinyBERT: 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
A Deep Relevance Matching Model for Ad-hoc Retrieval
Guo, J., Fan, Y., Ai, Q., & Croft, W. B. (2016) · Proceedings of the 25th ACM International on Conference on Information and Knowledge Management · doi:10.1145/2983323.2983769 · Cited by 806 · cited 1×
“Over the past few years, the Information Retrieval (IR) community has witnessed the introduction of a host of neural ranking models, including DRMM [8], KNRM [5, 36], and Duet [21, 23].”
BackgroundNogueira, R., & Cho, K. (2019) · arXiv · doi:10.48550/arxiv.1901.04085 · Cited by 615 · cited 9×
“Similar to [25], we fine-tune all ColBERT models with learning rate 3 × 10 − 6 with a batch size 32.”
BackgroundMethodInfluential
Research works printed but not linked: 14
- Zhuyun Dai and Jamie Callan . 2019 a. Context-Aware Sentence/Passage Term Importance Estimation For First Stage Retrieval. arXiv preprint arXiv:1910.10687 ( 2019 ). Zhuyun Dai and Jamie Callan. 2019 a. Context-Aware Sentence/Passage Term Importance Estimation For First Stage Retrieval. arXiv preprint arXiv:1910.10687 (2019). · arXiv
- Zhuyun Dai and Jamie Callan . 2019 b. Deeper Text Understanding for IR with Contextual Neural Language Modeling. arXiv preprint arXiv:1905.09217 ( 2019 ). Zhuyun Dai and Jamie Callan. 2019 b. Deeper Text Understanding for IR with Contextual Neural Language Modeling. arXiv preprint arXiv:1905.09217 (2019). · arXiv
- Jiafeng Guo , Yixing Fan , Liang Pang , Liu Yang , Qingyao Ai , Hamed Zamani , Chen Wu , W Bruce Croft , and Xueqi Cheng . 2019. A deep look into neural ranking models for information retrieval. arXiv preprint arXiv:1903.06902 ( 2019 ). Jiafeng Guo, Yixing Fan, Liang Pang, Liu Yang, Qingyao Ai, Hamed Zamani, Chen Wu, W Bruce Croft, and Xueqi Cheng. 2019. A deep look into neural ranking models for information retrieval. arXiv preprint arXiv:1903.06902 (2019). · arXiv
- Sebastian Hofst"atter and Allan Hanbury . 2019. Let's measure run time! Extending the IR replicability infrastructure to include performance aspects. arXiv preprint arXiv:1907.04614 ( 2019 ). Sebastian Hofst"atter and Allan Hanbury. 2019. Let's measure run time! Extending the IR replicability infrastructure to include performance aspects. arXiv preprint arXiv:1907.04614 (2019). · arXiv
- Sebastian Hofst"atter, Markus Zlabinger , and Allan Hanbury . 2019 b. TU Wien@ TREC Deep Learning'19--Simple Contextualization for Re-ranking. arXiv preprint arXiv:1912.01385 ( 2019 ). Sebastian Hofst"atter, Markus Zlabinger, and Allan Hanbury. 2019 b. TU Wien@ TREC Deep Learning'19--Simple Contextualization for Re-ranking. arXiv preprint arXiv:1912.01385 (2019). · arXiv
- Diederik P Kingma and Jimmy Ba . 2014 . Adam: A method for stochastic optimization. arXiv preprint arXiv:1412.6980 (2014). Diederik P Kingma and Jimmy Ba. 2014. Adam: A method for stochastic optimization. arXiv preprint arXiv:1412.6980 (2014). · arXiv
- Bhaskar Mitra and Nick Craswell . 2019. An Updated Duet Model for Passage Re-ranking. arXiv preprint arXiv:1903.07666 ( 2019 ). Bhaskar Mitra and Nick Craswell. 2019. An Updated Duet Model for Passage Re-ranking. arXiv preprint arXiv:1903.07666 (2019). · arXiv
- Bhaskar Mitra , Corby Rosset , David Hawking , Nick Craswell , Fernando Diaz , and Emine Yilmaz . 2019. Incorporating query term independence assumption for efficient retrieval and ranking using deep neural networks. arXiv preprint arXiv:1907.03693 ( 2019 ). Bhaskar Mitra, Corby Rosset, David Hawking, Nick Craswell, Fernando Diaz, and Emine Yilmaz. 2019. Incorporating query term independence assumption for efficient retrieval and ranking using deep neural networks. arXiv preprint arXiv:1907.03693 (2019). · arXiv
- Rodrigo Nogueira , Wei Yang , Kyunghyun Cho , and Jimmy Lin . 2019 b. Multi-Stage Document Ranking with BERT. arXiv preprint arXiv:1910.14424 ( 2019 ). Rodrigo Nogueira, Wei Yang, Kyunghyun Cho, and Jimmy Lin. 2019 b. Multi-Stage Document Ranking with BERT. arXiv preprint arXiv:1910.14424 (2019). · arXiv
- Rodrigo Nogueira , Wei Yang , Jimmy Lin , and Kyunghyun Cho . 2019 c. Document Expansion by Query Prediction. arXiv preprint arXiv:1904.08375 ( 2019 ). Rodrigo Nogueira, Wei Yang, Jimmy Lin, and Kyunghyun Cho. 2019 c. Document Expansion by Query Prediction. arXiv preprint arXiv:1904.08375 (2019). · arXiv
- Matthew E Peters , Mark Neumann , Mohit Iyyer , Matt Gardner , Christopher Clark , Kenton Lee , and Luke Zettlemoyer . 2018. Deep contextualized word representations. arXiv preprint arXiv:1802.05365 ( 2018 ). Matthew E Peters, Mark Neumann, Mohit Iyyer, Matt Gardner, Christopher Clark, Kenton Lee, and Luke Zettlemoyer. 2018. Deep contextualized word representations. arXiv preprint arXiv:1802.05365 (2018). · arXiv
- Yifan Qiao , Chenyan Xiong , Zhenghao Liu , and Zhiyuan Liu . 2019. Understanding the Behaviors of BERT in Ranking. arXiv preprint arXiv:1904.07531 ( 2019 ). Yifan Qiao, Chenyan Xiong, Zhenghao Liu, and Zhiyuan Liu. 2019. Understanding the Behaviors of BERT in Ranking. arXiv preprint arXiv:1904.07531 (2019). · arXiv
- Colin Raffel , Noam Shazeer , Adam Roberts , Katherine Lee , Sharan Narang , Michael Matena , Yanqi Zhou , Wei Li , and Peter J Liu . 2019. Exploring the limits of transfer learning with a unified text-to-text transformer. arXiv preprint arXiv:1910.10683 ( 2019 ). Colin Raffel, Noam Shazeer, Adam Roberts, Katherine Lee, Sharan Narang, Michael Matena, Yanqi Zhou, Wei Li, and Peter J Liu. 2019. Exploring the limits of transfer learning with a unified text-to-text transformer. arXiv preprint arXiv:1910.10683 (2019). · arXiv
- Raphael Tang , Yao Lu , Linqing Liu , Lili Mou , Olga Vechtomova , and Jimmy Lin . 2019. Distilling task-specific knowledge from BERT into simple neural networks. arXiv preprint arXiv:1903.12136 ( 2019 ). Raphael Tang, Yao Lu, Linqing Liu, Lili Mou, Olga Vechtomova, and Jimmy Lin. 2019. Distilling task-specific knowledge from BERT into simple neural networks. arXiv preprint arXiv:1903.12136 (2019). · arXiv
Cited by
Items 801–900 of 1,023
CoverageBench: Evaluating Information Coverage across Tasks and Domains
Samuel, S., Yates, A., Lawrie, D., Soboroff, I., Adriaanse, T., Van Durme, B., & Yang, E. (2026) · Proceedings of the 49th International ACM SIGIR Conference on Research and Development in Information Retrieval · doi:10.1145/3805712.3808625
A Comparative Study of Chunking Strategies for Retrieval-augmented Generation
Yuvzhenko, D., & Stirenko, S. (2026) · Cybernetics and Systems Analysis · doi:10.1007/s10559-026-00874-3 · Cited by 1
Mala, C., Gezici, G., & Giannotti, F. (2026) · Machine-mediated learning · doi:10.1007/s10994-026-07121-y
Tian, Y., Li, W., Chen, X., Brook, M., Brubaker, M., Liljedahl, A., & Baral, C. R. (2026) · ACM Transactions on Spatial Algorithms and Systems · doi:10.1145/3828544 · Cited by 1
Diaz-Santos, A., & Socha, K. (2026) · ACM SIGKDD Conference on Knowledge Discovery and Data Mining (KDD) · doi:10.1145/3770855.3818313 · Cited by 1
BBC: Improving Large- k Approximate Nearest Neighbor Search with a Bucket-Based Result Collector
Yin, Z., Cong, G., Zeng, K., Zhu, J., & Cui, B. (2026) · Proceedings of the VLDB Endowment · doi:10.14778/3836663.3836702 · Cited by 3
Scalability Analysis of Cross-modal Text-to-Image Pipelines
Singh, S. S., Barretto, Z., Khanijo, B., Sahu, A., & Ramanath, M. (2026) · ACM International Conference on Information and Knowledge Management (CIKM) · doi:10.1145/3799682.3840752 · Cited by 1
Designing production-friendly machine learning
Zaharia, M. (2021) · Proceedings of the VLDB Endowment · doi:10.14778/3484224.3484241 · Cited by 1
Self-citationYu, L., Jiang, D., Chen, K., & Shou, L. (2022) · International ACM SIGIR Conference on Research and Development in Information Retrieval (SIGIR) · doi:10.1145/3477495.3531664 · Cited by 1
KnowReQA: A Knowledge-aware Retrieval Question Answering System
Wang, C., Bai, J., Zhang, X., Yan, C., Ouyang, Y., Rong, W., & Xiong, Z. (2022) · Lecture notes in computer science · doi:10.1007/978-3-031-10983-6_54 · Cited by 1
A Semantic Search Framework for Similar Audit Issue Recommendation in Financial Industry
Zhang, C., Song, C., Agarwal, S., Wu, H., Zhang, X., & Lu, J. J. (2023) · ACM International Conference on Web Search and Data Mining (WSDM) · doi:10.1145/3539597.3573040 · Cited by 1
CAMVR: Context-Adaptive Multi-View Representation Learning for Dense Retrieval
Liao, Z., Hou, X., Lou, D., Chen, Y., & Zhang, N. (2023) · International Joint Conference on Neural Networks (IJCNN) · doi:10.1109/ijcnn54540.2023.10192020 · Cited by 1
“In ColBERT [23], [24], a relevance score is determined as the sum of the maximum similarity between the tokens of the query and document.”
BackgroundSample selecting method based on feature density for pest identification in smart agriculture
Li, X., Xiao, S., & Zhang, Z. (2023) · Sustainable Computing Informatics and Systems · doi:10.1016/j.suscom.2023.100904 · Cited by 3
How to Extract and Interact? Nested Siamese Text Matching with Interaction and Extraction
Zang, J., & Liu, H. (2023) · Lecture notes in computer science · doi:10.1007/978-3-031-44192-9_42 · Cited by 1
An Adaptive Mixup Hard Negative Sampling for Zero-Shot Entity Linking
Cai, S., Wu, X., Maimaiti, M., Chen, Y., Wang, Z., & Zheng, J. (2023) · Mathematics · doi:10.3390/math11204366 · Cited by 1
Style-Driven Multi-Perspective Relevance Mining Model for Hotspot Reprint Paragraph Prediction
Wang, L., Qian, H., Li, Q., Xu, D. J., & Zeng, D. D. (2023) · IEEE International Conference on Intelligence and Security Informatics (ISI) · doi:10.1109/isi58743.2023.10297268 · Cited by 1
“After that, the ColBERT[25] introduces the post-interaction architecture.”
BackgroundBuilding a Full-text Search Index Using "Transformer" Neural Network
Dobrynin, V., Abramovich, R., & Платонов, А. В. (2023) · IEEE International Conference on Application of Information and Communication Technologies (AICT) · doi:10.1109/aict59525.2023.10313200 · Cited by 3
Heterogeneous graph attention networks for passage retrieval
Albarede, L., Mulhem, P., Goeuriot, L., Marié, S., Pape, C. L., & Chardin-Segui, T. (2023) · Information Retrieval · doi:10.1007/s10791-023-09424-3 · Cited by 3
“AVG (Albarede et al. (2021)), two baseline models that do not consider context (fine-tuned ColBERT (Khattab and Zaharia (2020)), BM25 (Robert-son et al. (1995))) and an approach that focuses on the patent specific query generation step (Andersson et al. (2016)), namely Query - gen .”
Contextualizing and Expanding Conversational Queries without Supervision
Krasakis, A. M., Yates, A., & Kanoulas, E. (2023) · ACM Transactions on Information Systems · doi:10.1145/3632622 · Cited by 3
“In practice, we use ColBERT [37] as our query and document encoder f Enc .”
BackgroundMethodInfluentialOnce is Enough: A Light-Weight Cross-Attention for Fast Sentence Pair Modeling
Yang, Y., Qi, S., Liu, C., Wang, Q., Gao, C., & Xu, Z. (2023) · Conference on Empirical Methods in Natural Language Processing (EMNLP) · doi:10.18653/v1/2023.emnlp-main.168 · Cited by 1
Luo, W., Chen, Q., Zhang, Z., Huang, Z., & Cheng, G. (2023) · Findings of the Association for Computational Linguistics: EMNLP (Findings of EMNLP) · doi:10.18653/v1/2023.findings-emnlp.957 · Cited by 1
“We used five popular PLM-based methods for our experiments: monoBERT -large (Nogueira and Cho, 2019), 3 monoT5 -large (Nogueira et al., 2020), 4 coCondenser (Gao and Callan, 2022), 5 ColBERT -v2 (Khattab and Zaharia, 2020), 6 and ANCE (Xiong et al., 2021) 7 .”
MethodPersona-Coded Poly-Encoder: Persona-Guided Multi-Stream Conversational Sentence Scoring
Liu, J., Symons, C. T., & Vatsavai, R. R. (2023) · IEEE International Conference on Tools with Artificial Intelligence (ICTAI) · doi:10.1109/ictai59109.2023.00044 · Cited by 6
“We consider two strong retrieval-based methods with pre-fusion approaches as our baselines: PolyEncoder [13] and ColBERT [15].”
BackgroundMethodInfluentialKim, J., Park, S.-Y., Lee, J., Kim, J., & Kang, P. (2024) · Journal of Korean Institute of Industrial Engineers · doi:10.7232/jkiie.2024.50.1.036 · Cited by 1
Sharma, S. P. P. S. (2023) · Journal of Propulsion Technology · doi:10.52783/tjjpt.v44.i6.3408 · Cited by 1
“Reinforcement Learning : Some PIR systems use reinforcement learning to adapt and improve recommendations over time[12], [13].”
BackgroundDQNC2S: DQN-Based Cross-Stream Crisis Event Summarizer
Cambrin, D. R., Cagliero, L., & Garza, P. (2024) · Lecture notes in computer science · doi:10.1007/978-3-031-56063-7_34 · Cited by 1
“State-of-the-art approaches to the CrisisFACTS task rely on ColBERT[10] (ohmkiz [21]) or on the established Retrieve&Re-ranking approach (unicamp [16] and ohmkiz [21]).”
MethodLuo, M., Gokhale, T., Varshney, N., Yang, Y., & Baral, C. R. (2024) · Synthesis lectures on computer vision · doi:10.1007/978-3-031-57816-8_5 · Cited by 1
Clinical Trial Retrieval via Multi-grained Similarity Learning
Luo, J., Qian, C., Glass, L. M., & Ma, F. (2024) · International ACM SIGIR Conference on Research and Development in Information Retrieval (SIGIR) · doi:10.1145/3626772.3661366 · Cited by 2
Ambiguous Entity Oriented Targeted Document Detection
Shen, W., & Wen, H. (2024) · IEEE International Conference on Data Engineering (ICDE) · doi:10.1109/icde60146.2024.00072 · Cited by 2
Tell Me Why: Language Models Help Explain the Rationale Behind Internet Protocol Design
Bian, J., Welzl, M., Kutuzov, A., & Arefyev, N. (2024) · IEEE International Conference on Machine Learning for Communication and Networking (ICMLCN) · doi:10.1109/icmlcn59089.2024.10624781 · Cited by 4
“Some closely related tasks, e.g. Question Answering [32], [33], Information Retrieval [34], [35], Natural Language Inference [36], [37], Augment Mining [38], [39] have seen extensive exploration, as these are well-established research fields.”
BackgroundLatent Relations at Steady‐state with Associative Nets
Shabahang, K. D., Yim, H., & Dennis, S. J. (2024) · Cognitive Science · doi:10.1111/cogs.13494 · Cited by 2
Reasonable Sense of Direction: Making Course Recommendations Understandable with LLMs
Chun, H.-W., Ong, R. K., & Khong, A. W. H. (2024) · Midwest Symposium on Circuits and Systems · doi:10.1109/mwscas60917.2024.10658914 · Cited by 2
MARS: Matching Attribute-aware Representations for Text-based Sequential Recommendation
Kim, H., Kim, J., Choi, M., Lee, S., & Lee, J. (2024) · ACM International Conference on Information and Knowledge Management (CIKM) · doi:10.1145/3627673.3679960 · Cited by 3
“We employ maximum similarity (MaxSim) [16] to dynamically match each item attribute with different aspects of the user sequence, capturing user interest based on candidate item attributes.”
MethodGómez, E., Cuesta, H., Gkiokas, A., Gómez-Cañón, J. S., Porcaro, L., & Yesiler, F. (2024) · ACM eBooks · doi:10.1145/3674127.3702959 · Cited by 5
Baeza-Yates, R. A., Murgai, L., & Dudy, S. (2024) · ACM eBooks · doi:10.1145/3674127.3674138 · Cited by 1
Tong, Z., Luo, C., & Luo, T. (2024) · Conference proceedings/Conference proceedings - IEEE International Conference on Systems, Man, and Cybernetics · doi:10.1109/smc54092.2024.10831541 · Cited by 1
Performance Metrics Analysis for Deep Learning Models
Sharma, S., & Panda, S. P. (2024) · International Conference on Advances in Computation, Communication and Information Technology (ICAICCIT) · doi:10.1109/icaiccit64383.2024.10912181 · Cited by 1
How Vital Is the Jurisprudential Relevance: Law Article-Intervened Legal Case Retrieval and Matching
Xu, N., Wang, P., Liang, Z., Zhao, J., & Guan, X. (2025) · ACM Transactions on Information Systems · doi:10.1145/3725729 · Cited by 1
Enabling Low-Resource Language Retrieval: Establishing Baselines for Urdu MS MARCO
Butt, U., Varanasi, S., & Neumann, G. (2025) · Lecture notes in computer science · doi:10.1007/978-3-031-88717-8_21 · Cited by 1
Counterfactual Query Rewriting to Use Historical Relevance Feedback
Keller, J., Fröbe, M., Hendriksen, G., Alexander, D., Potthast, M., Hagen, M., & Schaer, P. (2025) · Lecture notes in computer science · doi:10.1007/978-3-031-88714-7_11 · Cited by 2
Heineking, S., Probst, J., Steinbach, D., Potthast, M., & Scells, H. (2025) · Lecture notes in computer science · doi:10.1007/978-3-031-88714-7_10 · Cited by 1
Enhancing Retrieval-Augmented Generation Systems by Text-Representing Centroid
Ruamsuk, Y., Mingkhwan, A., & Unger, H. (2024) · International Conference on Natural Language Processing and Information Retrieval (NLPIR) · doi:10.1145/3711542.3711558 · Cited by 1
“Furthermore,ourapproachfacilitatesmoreefficientandaccu-rateretrievalbyleveraginggraph-basedalgorithmsforinformation propagationandrelevancescoring[8].”
Dense Retrieval Systems with LLM-Based Query Expansion
Wang, Z., & Pei, Q. (2024) · IEEE/WIC/ACM International Conference on Web Intelligence and Intelligent Agent Technology (WI-IAT) · doi:10.1109/wi-iat62293.2024.00110 · Cited by 2
“ColBERT E2E [34] is an end-to-end retrieval method built on the ColBERT [5] dense retrieval model.”
KeyColBERT-PRF: Optimized Text Retrieval via Keyword Augmentation
Pan, M., Xie, Y., Zhou, S., Gao, M., Xiong, W., & Wang, M. (2024) · IEEE/WIC/ACM International Conference on Web Intelligence and Intelligent Agent Technology (WI-IAT) · doi:10.1109/wi-iat62293.2024.00122 · Cited by 1
“ColBERT [4], an optimization of BERT, demonstrates exceptional performance in document retrieval.”
Efficient Integration of ASR with Large Language Models to Enhance Video Search at Scale
Zhang, Q., Xiao, F., Li, T., Lin, L., Fang, H., Sun, H., Liu, R., Zhu, X., & Wang, J. (2025) · International World Wide Web Conference (WWW) · doi:10.1145/3701716.3715220 · Cited by 1
Wang, J., Zhang, F., Chen, X., He, P., Huang, E. A., & Huang, J. X. (2025) · ACM Transactions on Information Systems · doi:10.1145/3742423 · Cited by 3
“This framework uses the ColBERT [20] retriever for retrieval, aiming to rank highly relevant candidate documents as highly as possible during score calculation.”
A PROPOSED DISTRIBUTED ARCHITECTURE FOR SEARCHING SEMANTICALLY ON A LARGE DATASET OF HACKING NEWS
Do, N. L. (2025) · Journal of Science and Technique · doi:10.56651/lqdtu.jst.v14.n01.1040.ict · Cited by 1
Are Information Retrieval Approaches Good at Harmonising Longitudinal Surveys in Social Science?
Li, W. Y., Wang, Z., Johnson, J., & De, S. (2025) · International ACM SIGIR Conference on Research and Development in Information Retrieval (SIGIR) · doi:10.1145/3726302.3730164 · Cited by 1
“For efficiency, we follow the late interaction framework [8, 12] where representations are pre-computed offline with cosine distance (1 − _ ) as the similarity measure.”
Unsupervised Methods for Predicting Query Performance
Karimi, S., & Khodabakhsh, M. (2025) · Archives de l'Institut d'Hessarek. · doi:10.61186/jsdp.22.1.3 · Cited by 1
Robust Tree-based Learned Vector Index with Query-aware Repartitioning
Wei, W., Lian, D., Feng, Q., & Wu, Y. (2025) · ACM SIGKDD Conference on Knowledge Discovery and Data Mining (KDD) · doi:10.1145/3711896.3737112 · Cited by 1
Enhanced Arabic Text Retrieval with Attentive Relevance Scoring
Bekhouche, S. E., Benlamoudi, A., Bounab, Y., Dornaika, F., & Hadid, A. (2025) · Machine learning for signal processing ... · doi:10.1109/mlsp62443.2025.11204239 · Cited by 1
“While late-interaction models like ColBERT [9] address this by allowing token-level interactions, other neural ranking frameworks have long explored learned interaction layers on top of global embeddings to improve upon simple similarity measures [22, 23].”
Survey and Benchmarking of Retrieval Methods for Enterprise Retrieval-Augmented Generation
Ruparel, R., & Bussari, S. (2025) · International Conference on Computer and Applications (ICCA) · doi:10.1109/icca66035.2025.11430920 · Cited by 1
“Hybrid approaches that fuse sparse and dense signals, including late-interaction models such as Col-BERT/ColBERTv2 [12], [13], often yield the most robust overall performance [4], [5].”
Open Domain Question Answering Based on Retriever-Reader Architecture
Zheng, D., Yang, J., & Yong, B. (2021) · Lecture notes on data engineering and communications technologies · doi:10.1007/978-3-030-92632-8_68 · Cited by 1
Neural Pseudo-Relevance Feedback Models for Sparse and Dense Retrieval
Wang, X. (2022) · International ACM SIGIR Conference on Research and Development in Information Retrieval (SIGIR) · doi:10.1145/3477495.3531685 · Cited by 1
“In particular, we recently proposed to apply the pseudo-relevance feedback mechanism in a multiple representation dense retrieval setting [1], called ColBERT-PRF [4].”
MethodQ2R: A Query-to-Resolution System for Natural-Language Queries
Lim, S. H., & Wynter, L. (2022) · Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies (NAACL-HLT) · doi:10.18653/v1/2022.naacl-industry.39 · Cited by 1
VILT: Video Instructions Linking for Complex Tasks
Fischer, S., Gemmell, C., Mackie, I., & Dalton, J. (2022) · doi:10.1145/3552467.3554794 · Cited by 4
CDSM: C ascaded D eep S emantic M atching on Textual Graphs Leveraging Ad-hoc Neighbor Selection
Yao, J., Liu, Z., Yang, J., Dou, Z., Xie, X., & Wen, J.-R. (2022) · ACM Transactions on Intelligent Systems and Technology · doi:10.1145/3573204 · Cited by 1
“hanks to the increasing capacity of deep neural networks, a series of advanced document encoders have been proposed, especially those based on pre-trained language models [11, 15, 17].”
BackgroundA question answering model for electrical equipment standards based on subject object attention
Huayong, L., Ding, H., Hu, J., Lin, J., Chen, L., & Zhang, B. (2022) · International Conference on Systems and Informatics (ICSAI) · doi:10.1109/icsai57119.2022.10005504 · Cited by 1
ParaGraph: Mapping Wikidata Tail Entities to Wikipedia Paragraphs
Ostapuk, N., Difallah, D. E., & Cudré-Mauroux, P. (2022) · IEEE International Conference on Big Data (BigData) · doi:10.1109/bigdata55660.2022.10020207 · Cited by 2
“straightforward IR solution, we compare the performance of ParaGraph to the performance of BM25 as well as two stateof-the-art neural IR models: ColBERT [41] and ANCE [42].”
MethodImproving the Generalizability of the Dense Passage Retriever Using Generated Datasets
Rajapakse, T. C., & de Rijke, M. (2023) · Lecture notes in computer science · doi:10.1007/978-3-031-28238-6_7 · Cited by 8
“, ColBERT [14] and ANCE [28] with higher computational cost for training and/or retrieval.”
BackgroundMethodInfluentialLi, X., Zou, B., Dong, M., Yao, J., & Hong, Y. (2022) · IEEE International Conference on Tools with Artificial Intelligence (ICTAI) · doi:10.1109/ictai56018.2022.00053 · Cited by 1
“Recently, dense vector-based retrievers have been well developed [18], [19].”
BackgroundPseudo-Relevance for Enhancing Document Representation
Kim, J., Hwang, S.-W., Song, S. H., Ko, H., & Song, Y.-I. (2022) · Conference on Empirical Methods in Natural Language Processing (EMNLP) · doi:10.18653/v1/2022.emnlp-main.800 · Cited by 1
KALE: Using a K-Sparse Projector for Lexical Expansion
Borges, L., Martins, B., & Callan, J. (2023) · ACM SIGIR International Conference on Theory of Information Retrieval (ICTIR) · doi:10.1145/3578337.3605131 · Cited by 4
“In subsequent work, Lin et al. [21] further improved TCT-ColBERT with hard negative mining.”
MethodInfluentialResult Diversification for Legal case Retrieval
Zhang, R., Ai, Q., Wu, Y., Ma, Y., & Liu, Y. (2023) · Annual International ACM SIGIR Conference on Research and Development in Information Retrieval in the Asia Pacific Region (SIGIR-AP) · doi:10.1145/3624918.3625319 · Cited by 1
“The documents are then sorted based on their respective scores to produce the result list, adopting the methodology of ColBERT [19].”
MethodExplaining Semantic Text Similarity in Knowledge Graphs
Berlanga, R., & Soriano, M. A. (2023) · Lecture notes in computer science · doi:10.1007/978-3-031-49018-7_37 · Cited by 1
Empowering Language Model with Guided Knowledge Fusion for Biomedical Document Re-ranking
Gupta, D. P., & Demner‐Fushman, D. (2024) · Lecture notes in computer science · doi:10.1007/978-3-031-66538-7_25 · Cited by 3
MVAM: Multi-View Attention Method for Fine-Grained Image-Text Matching
Cui, W., Cheng, R., Guo, J., & Cheng, X. (2025) · Lecture notes in computer science · doi:10.1007/978-3-031-88711-6_11 · Cited by 1
Kalogeropoulos, N., Skamnelos, N., & Makris, C. H. (2025) · Knowledge and Information Systems · doi:10.1007/s10115-025-02470-x · Cited by 1
Composite Re-Ranking for Efficient Document Search with BERTPreprint
Yang, Y., Qiao, Y., Shao, J., Anand, M., Yan, X., & Yang, T. (2021) · arXiv (Cornell University) · doi:10.48550/arxiv.2103.06499 · Cited by 3
“For ColBert, we adopted the code released with the ColBERT paper 5 .”
BackgroundMethodInfluentialDistilling Transformers for Neural Cross-Domain SearchPreprint
Clement, C. B., Wu, C. H., Drain, D., & Sundaresan, N. (2021) · arXiv (Cornell University) · doi:10.48550/arxiv.2108.03322 · Cited by 2
“…(Matsubara et al., 2020), caching document-level representations for effi-cient queries (MacAvaney et al., 2020), novel architectures like ColBERT (Khattab and Zaharia, 2020) which allows query-document interactions while caching embeddings for efficient queries and MarkedBERT (Boualili et al.,…”
BackgroundUltra-High Dimensional Sparse Representations with Binarization for Efficient Text RetrievalPreprint
Jang, K.-R., Kang, J., Hong, G., Myaeng, S.-H., Park, J., Yoon, T., & Seo, H.-C. (2021) · arXiv (Cornell University) · doi:10.48550/arxiv.2104.07198 · Cited by 1
“Recent approaches to information retrieval (IR) and natural language processing (NLP) tasks adopt contextual language models (e.g., BERT (Devlin et al., 2019)), ameliorating both synonymy and pol-ysemy problems associated with words (Zhan et al., 2020; Khattab and Zaharia, 2020; Qu et al., 2021).”
BackgroundMethodInfluentialGeAR: Generation Augmented RetrievalPreprint
Liu, H., Huang, S., Liu, J., Zhan, Y., Sun, H., Deng, W., Sun, F., Wei, F., & Zhang, Q. (2025) · Qeios · doi:10.32388/gwq6lr · Cited by 1
Adapting Deep Neural Information Retrieval Models to Long Documents and New Domains
Li, M. (2023) · theses.fr (ABES) · doi:10.70675/b7c1a41fz0808z4c09zac20z6c9e21586cb0 · Cited by 1
Exploring Optimal Prefetching Sizes in RaLMSpec to Enhance Retrieval-Augmented Generation Efficiency
Liu, S. (2025) · Highlights in Science, Engineering and Technology · doi:10.54097/hhff9g78
“Khattab & Zaharia explored late interaction in ColBERT, demonstrating that optimized retrieval reduces latency but largely ignored prefetch-specific configurations [9].”
Lahmy, M., & Yozevitch, R. (2025) · arXiv · doi:10.48550/arxiv.2512.10787 · Cited by 1
“While dense retrievers like DPR [14] and late-interaction models like ColBERT [15] improve recall on single-hop queries, they often struggle with multi-hop reasoning, where the answer depends on composing information from multiple disjoint documents [20, 6].”
Spike Hijacking in Late-Interaction Retrieval
Suresh, K., Vatsa, T., King, T., Kadav, A., & Friedrich, M. (2026) · arXiv · doi:10.48550/arxiv.2604.05253
“Each query token contributes only its highest similarity over document patches [1].”
InfluentialSmartSearch: How Ranking Beats Structure for Conversational Memory Retrieval
Derehag, J., Calva, C., & Ghiurau, T. (2026) · arXiv · doi:10.48550/arxiv.2603.15599
“Dense retrieval: Embedding-based methods such as DPR (Karpukhin et al., 2020) and ColBERT (Khattab & Zaharia, 2020) bridge vocabulary gaps but are slower and require pre-computed indices.”
Synapse: Empowering LLM Agents with Episodic-Semantic Memory via Spreading Activation
Jiang, H., Chen, J., Pan, Y., Chen, L., You, W., Zhou, Y., Zhang, R., Abate, Y., & Liu, T. (2026) · Findings of the Association for Computational Linguistics: ACL 2026 · doi:10.18653/v1/2026.findings-acl.1108 · Cited by 2
Complementing Language Embeddings with Knowledge Bases for Specific Domains.
Tenti, P., Pasi, G., & Peñaloza, R. (2021)
Domain Adaptation for Medical Semantic Textual Similarity
Sun, J., & Li, S. (2021) · IEEE International Conference on Network Infrastructure and Digital Content (IC-NIDC) · doi:10.1109/ic-nidc54101.2021.9660484
Sun, H., Zhang, X., Wang, J., Li, K., Liao, J. B., & Su, Z. (2021) · IEEE International Conference on Power Data Science (ICPDS) · doi:10.1109/icpds54746.2021.9690073
Improving the Performance of Pre-trained Systems in Sentence Retrieval
Rughbeer, Y., Pillay, A. W., & Jembere, E. (2021) · International Conference on Information Fusion (FUSION) · doi:10.23919/fusion49465.2021.9627046
Another Look at Information Retrieval as Statistical Translation
Liu, Y., Hu, C., & Lin, J. (2022) · International ACM SIGIR Conference on Research and Development in Information Retrieval (SIGIR) · doi:10.1145/3477495.3531717 · Cited by 2
DKBERT: An Efficient and Effective Ranking Model via Weight Sharing Decoupling and Kernel Matching
Li, X., & Li, D. (2022) · International Conference on Artificial Intelligence and Big Data (ICAIBD) · doi:10.1109/icaibd55127.2022.9820107
Pre-training and evaluation of numeracy-oriented language model
Feng, F., Rui, X., Wang, W., Cao, Y., & Chua, T. (2021) · ACM International Conference on AI in Finance (ICAIF) · doi:10.1145/3490354.3494412
Training Framework for Passage Re-ranking based on Generating Adversarial Data
Bae, Y., & Lee, K.-J. (2022) · The Journal of Korean Institute of Information Technology · doi:10.14801/jkiit.2022.20.10.39
Mintamanis, J. C., & Mandala, R. (2022) · International Conference on Advanced Informatics: Concepts, Theory and Applications (ICAICTA) · doi:10.1109/icaicta56449.2022.9933005
“Research [3] examines sentence embedding in good information retrieval on the Natural Language Processing task.”
BackgroundMethodInfluentialCan a Machine Reading Comprehension Model Improve Ad-hoc Document Retrieval?
Usuha, K., Kato, M. P., & Fujita, S. (2022) · Lecture notes in computer science · doi:10.1007/978-3-031-21756-2_14
Liu, J. (2023) · The information retrieval series · doi:10.1007/978-3-031-23229-9_1
A Study of Term-Topic Embeddings for Ranking
Boualili, L., & Yates, A. (2023) · Lecture notes in computer science · doi:10.1007/978-3-031-28238-6_25
Improve retrieval of regional titles in streaming services with dense retrieval
Upadhyay, B., Khairnar, T., & Kotalwar, A. (2023) · International World Wide Web Conference (WWW) · doi:10.1145/3543873.3587619
“documents in the dense vector representation [10].”
MethodEnhancing BERT-based Passage Retriever with Word Recovery Method
Chen, H. Y., Qin, B., Yang, Y., & Zhang, J. (2022) · International Conference on Machine Learning, Big Data and Business Intelligence (MLBDBI) · doi:10.1109/mlbdbi58171.2022.00023
A Contrastive Framework for Improving Multi-turn Response Selection
Yang, Y., Qin, B., Chen, H., & Lan, G. (2022) · International Conference on Machine Learning, Big Data and Business Intelligence (MLBDBI) · doi:10.1109/mlbdbi58171.2022.00037
“Among them, the main examples include DPR [1], ColBERT [2], and RocketQA [3].”
BackgroundDeductive Additivity for Planning of Natural Language Proofs
Sprague, Z., Bostrom, K., Chaudhuri, S., & Durrett, G. (2023) · doi:10.18653/v1/2023.nlrse-1.11 · Cited by 1
Semantic Triple-Assisted Learning for Question Answering Passage Re-ranking
Nagumothu, D., Ofoghi, B., & Eklund, P. (2023) · Lecture notes in computer science · doi:10.1007/978-3-031-41682-8_16
LeCAR: Leveraging Context for Enhanced Automotive Specification Retrieval
Wu, K.-W., Hsu, T.-H., Huang, Y.-H., Chen, Y.-S., Wang, H.-L., Hsieh, B.-J., & Hsu, C.-H. (2023) · IEEE International Conference on Information Reuse and Integration (IRI) · doi:10.1109/iri58017.2023.00038
Discriminative Question Answering via Cascade Prompt Learning and Sentence Level Attention Mechanism
Xiaoguang, Y. U. A. N., Dai, C., Tian, Z., Xinyu, F. A. N., Yingyi, S. O. N. G., Zengwen, Y. U., Wang, P., & Ke, W. (2023) · IEICE Transactions on Information and Systems · doi:10.1587/transinf.2022edp7225
Candidate Generation for Entity Linking on Military Equipment
Hu, M., Zheng, S., & Chen, X. (2023) · International Conference on Artificial Intelligence and Computer Information Technology (AICIT) · doi:10.1109/aicit59054.2023.10277709
“This paper proposes a ColBERT-MIL-EL entity linking method based on the ColBERT[18].”
MethodStance-Aware Re-Ranking for Non-factual Comparative Queries
Reimer, J. H., Bondarenko, A., Fröbe, M., & Hagen, M. (2023) · Workshop on Argument Mining (ArgMining) · doi:10.18653/v1/2023.argmining-1.5
“Many participants of the task improved over a BM25 baseline (Robertson et al., 1994) by using neural (re-)ranking models like ColBERT (Khattab and Zaharia, 2020) or mono-and duoT5 (Pradeep et al., 2021), and by taking estimated argument quality into account.”
BackgroundPhan, D. T., & Le, H. T. (2023) · International Conference on Control, Automation and Information Sciences (ICCAIS) · doi:10.1109/iccais59597.2023.10382273
“Some approaches, like ColBERT [10], do not aggregate the model into a single sentence embedding but instead leverage a late interaction mechanism to fine-tune the model on question-answer data.”
MethodReport on the 1st Workshop on Generative Information Retrieval (Gen-IR 2023) at SIGIR 2023
Bénédict, G., Zhang, R., Metzler, D., Yates, A., Deffayet, R., Hager, P., & Jullien, S. (2023) · ACM SIGIR Forum · doi:10.1145/3642979.3642995 · Cited by 3
“• ColBert [Khattab and Zaharia, 2020] and T5 are the norm in retrieval, but can we use recent > 10B parameters autoregressive LLMs (e.g. Llama)?”
Method
Metadata
- Authors
- Khattab, O., & Zaharia, M.
- Affiliations
- Stanford University (US)
- Venue
- Proceedings of the 43rd International ACM SIGIR Conference on Research and Development in Information Retrieval
- Publisher
- ACM
- Published
- 2020-07-25
- Pages
- 39-48
- OpenAlex
- W3021397474
- Cited by
- 1,301
- References
- 21
- Citations by year
- 2026: 346 · 2025: 277 · 2024: 197 · 2023: 184 · 2022: 143 · 2021: 135 · 2020: 18
- FWCI
- 90.73
- Type
- conference paper
- Language
- en
- Open access
- closed
- Locations
- International ACM SIGIR Conference on Research and Development in Information Retrieval (SIGIR) (publishedVersion)
- Funders
- NSF (CNS-1651570)
- 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
- Keywords
- Colbert, BERT, document ranking, information retrieval, query-document similarity, efficient retrieval
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.