{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,9,19]],"date-time":"2026-09-19T05:02:24Z","timestamp":1789794144006,"version":"4.0.1"},"reference-count":54,"publisher":"Association for Natural Language Processing","issue":"3","content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["Journal of Natural Language Processing"],"published-print":{"date-parts":[[2026]]},"DOI":"10.5715\/jnlp.33.1484","type":"journal-article","created":{"date-parts":[[2026,9,14]],"date-time":"2026-09-14T22:14:17Z","timestamp":1789424057000},"page":"1484-1509","source":"Crossref","is-referenced-by-count":0,"title":["Token Pooling: Reducing the Footprint of Multi-Vector Retrieval with Minimal Performance Impact"],"prefix":"10.5715","volume":"33","author":[{"given":"Benjamin","family":"Clavi\u00e9","sequence":"first","affiliation":[{"name":"Mixedbread AI and National Institute of Informatics (NII), Tokyo, Japan. Parts of the work were carried out while affiliated with Answer.AI"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Antoine","family":"Chaffin","sequence":"additional","affiliation":[{"name":"LightOn, Paris, France"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Griffin","family":"Adams","sequence":"additional","affiliation":[{"name":"All work performed while at Answer.AI, San Francisco, USA."}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Makoto P.","family":"Kato","sequence":"additional","affiliation":[{"name":"University of Tsukuba and National Institute of Informatics, Tokyo, Japan."}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"3685","reference":[{"key":"1","doi-asserted-by":"crossref","unstructured":"Acquavia, A., Macdonald, C., and Tonellotto, N. (2023). \u201cStatic Pruning for Multi-representation Dense Retrieval.\u201d In <i>Proceedings of the ACM Symposium on Document Engineering 2023<\/i>, pp. 1\u201310.","DOI":"10.1145\/3573128.3604896"},{"key":"2","unstructured":"Bai, J., Bai, S., Yang, S., Wang, S., Tan, S., Wang, P., Lin, J., Zhou, C., and Zhou, J. (2023). \u201cQwen-VL: A Versatile Vision-Language Model for Understanding, Localization, Text Reading, and Beyond.\u201d <i>arXiv preprint arXiv:2308.12966<\/i>."},{"key":"3","unstructured":"Beyer, L., Steiner, A., Pinto, A. S., Kolesnikov, A., Wang, X., Salz, D., Neumann, M., Alabdulmohsin, I., Tschannen, M., Bugliarello, E., et al. (2024). \u201cPaliGemma: A Versatile 3B VLM for Transfer.\u201d <i>arXiv preprint arXiv:2407.07726<\/i>."},{"key":"4","unstructured":"Bolya, D., Fu, C.-Y., Dai, X., Zhang, P., Feichtenhofer, C., and Hoffman, J. (2022). \u201cToken Merging: Your ViT but Faster.\u201d <i>arXiv preprint arXiv:2210.09461<\/i>."},{"key":"5","doi-asserted-by":"crossref","unstructured":"Bolya, D. and Hoffman, J. (2023). \u201cToken Merging for Fast Stable Diffusion.\u201d In <i>Proceedings of the IEEE\/CVF Conference on Computer Vision and Pattern Recognition<\/i>, pp. 4599\u20134603.","DOI":"10.1109\/CVPRW59228.2023.00484"},{"key":"6","doi-asserted-by":"crossref","unstructured":"Boteva, V., Gholipour, D., Sokolov, A., and Riezler, S. (2016). \u201cA Full-text Learning to Rank Dataset for Medical Information Retrieval.\u201d In <i>Advances in Information Retrieval: 38th European Conference on IR Research, ECIR 2016, Padua, Italy, March 20\u201323, 2016. Proceedings 38<\/i>, pp. 716\u2013722. Springer.","DOI":"10.1007\/978-3-319-30671-1_58"},{"key":"7","doi-asserted-by":"crossref","unstructured":"Brunila, M. and LaViolette, J. (2022). \u201cWhat Company Do Words Keep? Revisiting the Distributional Semantics of JR Firth &amp; Zellig Harris.\u201d <i>arXiv preprint arXiv:2205.07750<\/i>.","DOI":"10.18653\/v1\/2022.naacl-main.327"},{"key":"8","doi-asserted-by":"crossref","unstructured":"Chaffin, A. and Sourty, R. (2025). \u201cPylate: Flexible Training and Retrieval for Late Interaction Models.\u201d <i>arXiv preprint arXiv:2508.03555, to be published at CIKM 2025<\/i>.","DOI":"10.1145\/3746252.3761608"},{"key":"9","doi-asserted-by":"crossref","unstructured":"Chen, Y., Yuan, J., Tian, Y., Geng, S., Li, X., Zhou, D., Metaxas, D. N., and Yang, H. (2023). \u201cRevisiting Multimodal Representation in Contrastive Learning: From Patch and Token Embeddings to Finite Discrete Tokens.\u201d In <i>Proceedings of the IEEE\/CVF Conference on Computer Vision and Pattern Recognition<\/i>, pp. 15095\u201315104.","DOI":"10.1109\/CVPR52729.2023.01449"},{"key":"10","unstructured":"Clavi\u00e9, B., Chaffin, A., and Adams, G. (2024). \u201cReducing the Footprint of Multi-Vector Retrieval with Minimal Performance Impact via Token Pooling.\u201d <i>arXiv preprint arXiv:2409.14683<\/i>."},{"key":"11","doi-asserted-by":"crossref","unstructured":"Cohan, A., Feldman, S., Beltagy, I., Downey, D., and Weld, D. S. (2020). \u201cSPECTER: Document-level Representation Learning using Citation-informed Transformers.\u201d In <i>Proceedings of the 58th Annual Meeting of the Association for Computational Linguistics<\/i>, pp. 2270\u20132282.","DOI":"10.18653\/v1\/2020.acl-main.207"},{"key":"12","doi-asserted-by":"crossref","unstructured":"Craswell, N., Mitra, B., Yilmaz, E., and Campos, D. (2021). \u201cOverview of the TREC 2020 Deep Learning Track.\u201d <i>arXiv preprint arXiv:2102.07662<\/i>.","DOI":"10.6028\/NIST.SP.1266.deep-overview"},{"key":"13","doi-asserted-by":"crossref","unstructured":"Craswell, N., Mitra, B., Yilmaz, E., Campos, D., and Voorhees, E. M. (2020). \u201cOverview of the TREC 2019 Deep Learning Track.\u201d <i>arXiv preprint arXiv:2003.07820<\/i>.","DOI":"10.6028\/NIST.SP.1266.deep-overview"},{"key":"14","unstructured":"de Souza P. Moreira, G., Ak, R., Xu, M., Holworthy, O., Schifferer, B., Yu, Z., Babakhin, Y., Osmulski, R., Cai, J., Chesler, R., Liu, B., and Oldridge, E. (2026). \u201cNemotron ColEmbed V2: Top-Performing Late Interaction Embedding Models for Visual Document Retrieval.\u201d In <i>The 1st Late Interaction Workshop (LIR) @ ECIR 2026<\/i>."},{"key":"15","unstructured":"Faysse, M., Sibille, H., Wu, T., Omrani, B., Viaud, G., HUDELOT, C., and Colombo, P. (2025). \u201cColPali: Efficient Document Retrieval with Vision Language Models.\u201d In <i>The 13th International Conference on Learning Representations<\/i>."},{"key":"16","unstructured":"Firth, J. (1957). \u201cA Synopsis of Linguistic Theory, 1930\u20131955.\u201d In <i>Studies in Linguistic Analysis<\/i>, pp. 10\u201332. The Philological Society."},{"key":"17","doi-asserted-by":"crossref","unstructured":"Formal, T., Piwowarski, B., and Clinchant, S. (2021). \u201cSPLADE: Sparse Lexical and Expansion Model for First Stage Ranking.\u201d In <i>Proceedings of the 44th International ACM SIGIR Conference on Research and Development in Information Retrieval<\/i>, pp. 2288\u20132292.","DOI":"10.1145\/3404835.3463098"},{"key":"18","doi-asserted-by":"crossref","unstructured":"He, S., Al-Darabsah, M., Nair, S., May, J., Agarwal, T., Yang, T., and Teo, C. H. (2025). \u201cToken Pruning Optimization for Efficient Multi-vector Dense Retrieval.\u201d In Hauff, C., Macdonald, C., Jannach, D., Kazai, G., Nardini, F. M., Pinelli, F., Silvestri, F., and Tonellotto, N. (Eds.), <i>Advances in Information Retrieval<\/i>, pp. 101\u2013115, Cham. Springer Nature Switzerland.","DOI":"10.1007\/978-3-031-88708-6_7"},{"key":"19","unstructured":"Huang, X. and Tan, K. M. (2025). \u201cBeyond Text: Unlocking True Multimodal, End-to-end RAG with Tomoro ColQwen3.\u201d Tomoro.ai Web Article."},{"key":"20","doi-asserted-by":"crossref","unstructured":"Jegou, H., Douze, M., and Schmid, C. (2010). \u201cProduct Quantization for Nearest Neighbor Search.\u201d <i>IEEE Transactions on Pattern Analysis and Machine Intelligence<\/i>, <b>33<\/b> (1), pp. 117\u2013128.","DOI":"10.1109\/TPAMI.2010.57"},{"key":"21","unstructured":"Jha, R., Zuo, C., Kriz, R., and Durme, B. V. (2026). \u201cA Brief Comparison of Training-Free Multi-Vector Sequence Compression Methods.\u201d In <i>The 1st Late Interaction Workshop (LIR) @ ECIR 2026<\/i>."},{"key":"22","doi-asserted-by":"crossref","unstructured":"Karpukhin, V., Oguz, B., Min, S., Lewis, P., Wu, L., Edunov, S., Chen, D., and Yih, W.-t. (2020). \u201cDense Passage Retrieval for Open-Domain Question Answering.\u201d In <i>Proceedings of the 2020 Conference on Empirical Methods in Natural Language Processing (EMNLP)<\/i>, pp. 6769\u20136781.","DOI":"10.18653\/v1\/2020.emnlp-main.550"},{"key":"23","doi-asserted-by":"crossref","unstructured":"Khattab, O. and Zaharia, M. (2020). \u201cColbert: Efficient and Effective Passage Search via Contextualized Late Interaction Over Bert.\u201d In <i>Proceedings of the 43rd International ACM SIGIR Conference on Research and Development in Information Retrieval<\/i>, pp. 39\u201348.","DOI":"10.1145\/3397271.3401075"},{"key":"24","doi-asserted-by":"crossref","unstructured":"Korenius, T., Laurikkala, J., Juhola, M., and J\u00e4rvelin, K. (2006). \u201cHierarchical Clustering of a Finnish Newspaper Article Collection with Graded Relevance Assessments.\u201d <i>Information Retrieval<\/i>, <b>9<\/b>, pp. 33\u201353.","DOI":"10.1007\/s10791-005-5720-6"},{"key":"25","doi-asserted-by":"crossref","unstructured":"Lassance, C., Maachou, M., Park, J., and Clinchant, S. (2022). \u201cLearned Token Pruning in Contextualized Late Interaction Over BERT (colbert).\u201d In <i>Proceedings of the 45th International ACM SIGIR Conference on Research and Development in Information Retrieval<\/i>, pp. 2232\u20132236.","DOI":"10.1145\/3477495.3531835"},{"key":"26","doi-asserted-by":"crossref","unstructured":"Lee, J., Dai, Z., Duddu, S. M. K., Lei, T., Naim, I., Chang, M.-W., and Zhao, V. (2024). \u201cRethinking the Role of Token Retrieval in Multi-vector Retrieval.\u201d <i>Advances in Neural Information Processing Systems<\/i>, <b>36<\/b>.","DOI":"10.52202\/075280-0677"},{"key":"27","unstructured":"Li, M., Zhang, Y., Long, D., Chen, K., Song, S., Bai, S., Yang, Z., Xie, P., Yang, A., Liu, D., Zhou, J., and Lin, J. (2026). \u201cQwen3-VL-Embedding and Qwen3-VL-Reranker: A Unified Framework for State-of-the-Art Multimodal Retrieval and Ranking.\u201d <i>arXiv preprint arXiv:2601.04720<\/i>."},{"key":"28","doi-asserted-by":"crossref","unstructured":"Lupart, S., Formal, T., and Clinchant, S. (2023). \u201cMs-shift: An Analysis of ms Marco Distribution Shifts on Neural Retrieval.\u201d In <i>European Conference on Information Retrieval<\/i>, pp. 636\u2013652. Springer.","DOI":"10.1007\/978-3-031-28244-7_40"},{"key":"29","doi-asserted-by":"crossref","unstructured":"Ma, X., Lin, S.-C., Li, M., Chen, W., and Lin, J. (2024). \u201cUnifying Multimodal Retrieval via Document Screenshot Embedding.\u201d In Al-Onaizan, Y., Bansal, M., and Chen, Y.-N. (Eds.), <i>Proceedings of the 2024 Conference on Empirical Methods in Natural Language Processing<\/i>, pp. 6492\u20136505, Miami, Florida, USA. Association for Computational Linguistics.","DOI":"10.18653\/v1\/2024.emnlp-main.373"},{"key":"30","doi-asserted-by":"crossref","unstructured":"Ma, Y., Li, J., Zang, Y., Wu, X., Dong, X., Zhang, P., Cao, Y., Duan, H., Wang, J., Cao, Y., and Sun, A. (2025). \u201cTowards Storage-Efficient Visual Document Retrieval: An Empirical Study on Reducing Patch-Level Embeddings.\u201d <i>arXiv preprint arXiv:2506.04997<\/i>.","DOI":"10.18653\/v1\/2025.findings-acl.1003"},{"key":"31","doi-asserted-by":"crossref","unstructured":"MacAvaney, S., Mallia, A., and Tonellotto, N. (2025). \u201cEfficient Constant-Space Multi-vector Retrieval.\u201d In <i>European Conference on Information Retrieval<\/i>, pp. 237\u2013245. Springer.","DOI":"10.1007\/978-3-031-88714-7_22"},{"key":"32","unstructured":"Mac\u00e9, Q., Loison, A., and Faysse, M. (2025). \u201cViDoRe Benchmark V2: Raising the Bar for Visual Retrieval.\u201d <i>arXiv preprint arXiv:2505.17166<\/i>."},{"key":"33","unstructured":"MacQueen, J. et al. (1967). \u201cSome Methods for Classification and Analysis of Multivariate Observations.\u201d In <i>Proceedings of the 5th Berkeley Symposium on Mathematical Statistics and Probability<\/i>, Vol. 1, pp. 281\u2013297. Oakland, CA, USA."},{"key":"34","doi-asserted-by":"crossref","unstructured":"Maia, M., Handschuh, S., Freitas, A., Davis, B., McDermott, R., Zarrouk, M., and Balahur, A. (2018). \u201cWww\u201918 Open Challenge: Financial Opinion Mining and Question Answering.\u201d In <i>Companion Proceedings of the the Web Conference 2018<\/i>, pp. 1941\u20131942.","DOI":"10.1145\/3184558.3192301"},{"key":"35","doi-asserted-by":"crossref","unstructured":"Malkov, Y. A. and Yashunin, D. A. (2018). \u201cEfficient and Robust Approximate Nearest Neighbor Search using Hierarchical Navigable Small World Graphs.\u201d <i>IEEE Transactions on Pattern Analysis and Machine Intelligence<\/i>, <b>42<\/b> (4), pp. 824\u2013836.","DOI":"10.1109\/TPAMI.2018.2889473"},{"key":"36","doi-asserted-by":"crossref","unstructured":"Marin, D., Chang, J.-H. R., Ranjan, A., Prabhu, A., Rastegari, M., and Tuzel, O. (2023). \u201cToken Pooling in Vision Transformers for Image Classification.\u201d In <i>Proceedings of the IEEE\/CVF Winter Conference on Applications of Computer Vision<\/i>, pp. 12\u201321.","DOI":"10.1109\/WACV56688.2023.00010"},{"key":"37","doi-asserted-by":"crossref","unstructured":"Murtagh, F. and Contreras, P. (2012). \u201cAlgorithms for Hierarchical Clustering: An Overview.\u201d <i>Wiley Interdisciplinary Reviews: Data Mining and Knowledge Discovery<\/i>, <b>2<\/b> (1), pp. 86\u201397.","DOI":"10.1002\/widm.53"},{"key":"38","unstructured":"Nguyen, T., Rosenberg, M., Song, X., Gao, J., Tiwary, S., Majumder, R., and Deng, L. (2016). \u201cMS MARCO: A Human-generated Machine Reading Comprehension Dataset.\u201d <i>arXiv preprint arXiv:1611.09268<\/i>."},{"key":"39","unstructured":"Paszke, A., Gross, S., Massa, F., Lerer, A., Bradbury, J., Chanan, G., Killeen, T., Lin, Z., Gimelshein, N., Antiga, L., et al. (2019). \u201cPytorch: An Imperative Style, High-performance Deep Learning Library.\u201d <i>Advances in Neural Information Processing Systems<\/i>, <b>32<\/b>."},{"key":"40","doi-asserted-by":"crossref","unstructured":"Salih, N. M. and Jacksi, K. (2020). \u201cSemantic Document Clustering using K-means Algorithm and Ward\u2019s Method.\u201d In <i>2020 International Conference on Advanced Science and Engineering (ICOASE)<\/i>, pp. 1\u20136. IEEE.","DOI":"10.1109\/ICOASE51841.2020.9436588"},{"key":"41","doi-asserted-by":"crossref","unstructured":"Santhanam, K., Khattab, O., Potts, C., and Zaharia, M. (2022a). \u201cPLAID: An Efficient Engine for Late Interaction Retrieval.\u201d In <i>Proceedings of the 31st ACM International Conference on Information &amp; Knowledge Management<\/i>, pp. 1747\u20131756.","DOI":"10.1145\/3511808.3557325"},{"key":"42","doi-asserted-by":"crossref","unstructured":"Santhanam, K., Khattab, O., Saad-Falcon, J., Potts, C., and Zaharia, M. (2022b). \u201cColBERTv2: Effective and Efficient Retrieval via Lightweight Late Interaction.\u201d In <i>Proceedings of the 2022 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies<\/i>, pp. 3715\u20133734.","DOI":"10.18653\/v1\/2022.naacl-main.272"},{"key":"43","doi-asserted-by":"crossref","unstructured":"Thakur, N., Bonifacio, L., Fr\u00f6be, M., Bondarenko, A., Kamalloo, E., Potthast, M., Hagen, M., and Lin, J. (2024). \u201cSystematic Evaluation of Neural Retrieval Models on the Touch\u00e9 2020 Argument Retrieval Subset of BEIR.\u201d In <i>Proceedings of the 47th International ACM SIGIR Conference on Research and Development in Information Retrieval<\/i>, pp. 1420\u20131430.","DOI":"10.1145\/3626772.3657861"},{"key":"44","unstructured":"Thakur, N., Reimers, N., R\u00fcckl\u00e9, A., Srivastava, A., and Gurevych, I. (2021). \u201cBEIR: A Heterogeneous Benchmark for Zero-shot Evaluation of Information Retrieval Models.\u201d <i>arXiv preprint arXiv:2104.08663<\/i>."},{"key":"45","unstructured":"Veneroso, J., Jayaram, R., Rao, J., \u00c1brego, G. H., Hadian, M., and Cer, D. (2025). \u201cCRISP: Clustering Multi-Vector Representations for Denoising and Pruning.\u201d <i>arXiv preprint arXiv:2505.11471<\/i>."},{"key":"46","doi-asserted-by":"crossref","unstructured":"Virtanen, P., Gommers, R., Oliphant, T. E., Haberland, M., Reddy, T., Cournapeau, D., Burovski, E., Peterson, P., Weckesser, W., Bright, J., et al. (2020). \u201cSciPy 1.0: Fundamental Algorithms for Scientific Computing in Python.\u201d <i>Nature Methods<\/i>, <b>17<\/b> (3), pp. 261\u2013272.","DOI":"10.1038\/s41592-020-0772-5"},{"key":"47","unstructured":"Voyager Team, S. (2024). \u201cVoyager: An Approximate Nearest-Neighbor Search Library for Python and Java.\u201d https:\/\/github.com\/spotify\/voyager."},{"key":"48","doi-asserted-by":"crossref","unstructured":"Wadden, D., Lin, S., Lo, K., Wang, L. L., van Zuylen, M., Cohan, A., and Hajishirzi, H. (2020). \u201cFact or Fiction: Verifying Scientific Claims.\u201d In <i>Proceedings of the 2020 Conference on Empirical Methods in Natural Language Processing (EMNLP)<\/i>, pp. 7534\u20137550.","DOI":"10.18653\/v1\/2020.emnlp-main.609"},{"key":"49","doi-asserted-by":"crossref","unstructured":"Ward Jr, J. H. (1963). \u201cHierarchical Grouping to Optimize an Objective Function.\u201d <i>Journal of the American Statistical Association<\/i>, <b>58<\/b> (301), pp. 236\u2013244.","DOI":"10.1080\/01621459.1963.10500845"},{"key":"50","unstructured":"Weller, O., Boratko, M., Naim, I., and Lee, J. (2026). \u201cOn the Theoretical Limitations of Embedding-Based Retrieval.\u201d In <i>The 14th International Conference on Learning Representations<\/i>."},{"key":"51","doi-asserted-by":"crossref","unstructured":"Yates, A., Nogueira, R., and Lin, J. (2021). \u201cPretrained Transformers for Text Ranking: BERT and Beyond.\u201d In <i>Proceedings of the 14th ACM International Conference on Web Search and Data Mining<\/i>, pp. 1154\u20131156.","DOI":"10.1145\/3437963.3441667"},{"key":"52","doi-asserted-by":"crossref","unstructured":"Zhan, J., Xie, X., Mao, J., Liu, Y., Guo, J., Zhang, M., and Ma, S. (2022). \u201cEvaluating Interpolation and Extrapolation Performance of Neural Retrieval Models.\u201d In <i>Proceedings of the 31st ACM International Conference on Information &amp; Knowledge Management<\/i>, pp. 2486\u20132496.","DOI":"10.1145\/3511808.3557312"},{"key":"53","doi-asserted-by":"crossref","unstructured":"Zobel, J. and Moffat, A. (2006). \u201cInverted Files for Text Search Engines.\u201d <i>ACM Computing Surveys (CSUR)<\/i>, <b>38<\/b> (2), pp. 6\u2013es.","DOI":"10.1145\/1132956.1132959"},{"key":"54","doi-asserted-by":"crossref","unstructured":"Zong, Y. and Piwowarski, B. (2025). \u201cTowards Lossless Token Pruning in Late-Interaction Retrieval Models.\u201d In <i>Proceedings of the 48th International ACM SIGIR Conference on Research and Development in Information Retrieval<\/i>, SIGIR \u201925, pp. 2407\u20132417, New York, NY, USA. Association for Computing Machinery.","DOI":"10.1145\/3726302.3730100"}],"container-title":["Journal of Natural Language Processing"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/www.jstage.jst.go.jp\/article\/jnlp\/33\/3\/33_1484\/_pdf","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,9,19]],"date-time":"2026-09-19T04:30:21Z","timestamp":1789792221000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.jstage.jst.go.jp\/article\/jnlp\/33\/3\/33_1484\/_article\/-char\/ja\/"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026]]},"references-count":54,"journal-issue":{"issue":"3","published-print":{"date-parts":[[2026]]}},"URL":"https:\/\/doi.org\/10.5715\/jnlp.33.1484","relation":{},"ISSN":["1340-7619","2185-8314"],"issn-type":[{"value":"1340-7619","type":"print"},{"value":"2185-8314","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026]]}}}