Librarian Bot. I found the following papers similar to this paper.

The following papers were recommended by the Semantic Scholar API

mGTE: Generalized Long-Context Text Representation and Reranking Models for Multilingual Text Retrieval (2024)
Mistral-SPLADE: LLMs for better Learned Sparse Retrieval (2024)
Large Language Models as Foundations for Next-Gen Dense Retrieval: A Comprehensive Empirical Assessment (2024)
Mamba Retriever: Utilizing Mamba for Effective and Efficient Dense Retrieval (2024)
NV-Retriever: Improving text embedding models with effective hard-negative mining (2024)

Please give a thumbs up to this comment if you found it helpful!

If you want recommendations for any Paper on Hugging Face checkout this Space

You can directly ask Librarian Bot for paper recommendations by tagging it in a comment: \n\n@librarian-bot\n\t recommend

\n","updatedAt":"2024-09-03T01:34:16.120Z","author":{"_id":"63d3e0e8ff1384ce6c5dd17d","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/1674830754237-63d3e0e8ff1384ce6c5dd17d.jpeg","fullname":"Librarian Bot (Bot)","name":"librarian-bot","type":"user","isPro":false,"isHf":false,"isHfAdmin":false,"isMod":false,"followerCount":318,"isUserFollowing":false}},"numEdits":0,"identifiedLanguage":{"language":"en","probability":0.7307335138320923},"editors":["librarian-bot"],"editorAvatarUrls":["https://cdn-avatars.huggingface.co/v1/production/uploads/1674830754237-63d3e0e8ff1384ce6c5dd17d.jpeg"],"reactions":[],"isReport":false}}],"primaryEmailConfirmed":false,"paper":{"id":"2408.16672","authors":[{"_id":"66d585b7092cc13d0f961d44","name":"Rohan Jha","hidden":false},{"_id":"66d585b7092cc13d0f961d45","user":{"_id":"63491dc83d8dc83a55cb749c","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/63491dc83d8dc83a55cb749c/IoqJrOIaEnYO_S7si4KGp.jpeg","isPro":false,"fullname":"Bo Wang","user":"bwang0911","type":"user"},"name":"Bo Wang","status":"extracted_confirmed","statusLastChangedAt":"2024-10-02T07:11:25.862Z","hidden":false},{"_id":"66d585b7092cc13d0f961d46","user":{"_id":"6476ff2699a5ce743ccea3fc","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/6476ff2699a5ce743ccea3fc/zmFmF8tXXDaAGcl8RYiRr.jpeg","isPro":false,"fullname":"Michael Günther","user":"michael-guenther","type":"user"},"name":"Michael Günther","status":"admin_assigned","statusLastChangedAt":"2024-09-02T09:35:08.111Z","hidden":false},{"_id":"66d585b7092cc13d0f961d47","user":{"_id":"64c23f6d569648a60737eddb","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/64c23f6d569648a60737eddb/iZq7bp-yYaGl5VBVoN5Dg.jpeg","isPro":false,"fullname":"Saba Sturua","user":"jupyterjazz","type":"user"},"name":"Saba Sturua","status":"admin_assigned","statusLastChangedAt":"2024-09-02T09:35:17.536Z","hidden":false},{"_id":"66d585b7092cc13d0f961d48","user":{"_id":"64d22f33032a420d1863b6ea","avatarUrl":"/avatars/ed3eaf4bab70dd6ab9a2b67b5928e4fb.svg","isPro":false,"fullname":"Mohammad Kalim Akram","user":"makram93","type":"user"},"name":"Mohammad Kalim Akram","status":"admin_assigned","statusLastChangedAt":"2024-09-02T09:35:23.144Z","hidden":false},{"_id":"66d585b7092cc13d0f961d49","user":{"_id":"603763514de52ff951d89793","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/603763514de52ff951d89793/n-QouGYg7oE5QeDaAb3Ns.png","isPro":false,"fullname":"Han Xiao","user":"hanxiao","type":"user"},"name":"Han Xiao","status":"admin_assigned","statusLastChangedAt":"2024-09-02T09:35:38.480Z","hidden":false}],"publishedAt":"2024-08-29T16:21:00.000Z","submittedOnDailyAt":"2024-09-02T08:00:33.428Z","title":"Jina-ColBERT-v2: A General-Purpose Multilingual Late Interaction\n Retriever","submittedOnDailyBy":{"_id":"62716952bcef985363db8485","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/62716952bcef985363db8485/zJPPo5xlwZRJdEuwYsYKp.jpeg","isPro":true,"fullname":"JB D.","user":"IAMJB","type":"user"},"summary":"Multi-vector dense models, such as ColBERT, have proven highly effective in\ninformation retrieval. ColBERT's late interaction scoring approximates the\njoint query-document attention seen in cross-encoders while maintaining\ninference efficiency closer to traditional dense retrieval models, thanks to\nits bi-encoder architecture and recent optimizations in indexing and search. In\nthis paper, we introduce several improvements to the ColBERT model architecture\nand training pipeline, leveraging techniques successful in the more established\nsingle-vector embedding model paradigm, particularly those suited for\nheterogeneous multilingual data. Our new model, Jina-ColBERT-v2, demonstrates\nstrong performance across a range of English and multilingual retrieval tasks,\nwhile also cutting storage requirements by up to 50% compared to previous\nmodels.","upvotes":9,"discussionId":"66d585b8092cc13d0f961d6e","ai_summary":"Jina-ColBERT-v2 enhances ColBERT's performance and reduces storage requirements through improvements to its architecture and training pipeline, benefiting multilingual retrieval tasks.","ai_keywords":["multi-vector dense models","ColBERT","late interaction scoring","joint query-document attention","cross-encoders","bi-encoder architecture","indexing","search","single-vector embedding","Jina-ColBERT-v2","heterogeneous multilingual data"],"organization":{"_id":"63563e0c2d14fcd7d83743cf","name":"jinaai","fullname":"Jina AI","avatar":"https://cdn-uploads.huggingface.co/production/uploads/603763514de52ff951d89793/wD54VbAHHyHop3uYlJKl4.png"}},"canReadDatabase":false,"canManagePapers":false,"canSubmit":false,"hasHfLevelAccess":false,"upvoted":false,"upvoters":[{"_id":"62716952bcef985363db8485","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/62716952bcef985363db8485/zJPPo5xlwZRJdEuwYsYKp.jpeg","isPro":true,"fullname":"JB D.","user":"IAMJB","type":"user"},{"_id":"648eb1eb59c4e5c87dc116e0","avatarUrl":"/avatars/c636cea39c2c0937f01398c94ead5dad.svg","isPro":false,"fullname":"fdsqefsgergd","user":"T-representer","type":"user"},{"_id":"620783f24e28382272337ba4","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/620783f24e28382272337ba4/zkUveQPNiDfYjgGhuFErj.jpeg","isPro":false,"fullname":"GuoLiangTang","user":"Tommy930","type":"user"},{"_id":"648e72a866dcba8b5aaecbdc","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/noauth/UKR1KF7s0OWCFQzsTrcuX.jpeg","isPro":false,"fullname":"Sergey Bratchikov","user":"hivaze","type":"user"},{"_id":"64c23f6d569648a60737eddb","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/64c23f6d569648a60737eddb/iZq7bp-yYaGl5VBVoN5Dg.jpeg","isPro":false,"fullname":"Saba Sturua","user":"jupyterjazz","type":"user"},{"_id":"639c2b60db7c5f35003f4168","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/639c2b60db7c5f35003f4168/wTIY98vCcpyPaR3zGz7hd.jpeg","isPro":false,"fullname":"李浩","user":"lihaocruiser","type":"user"},{"_id":"64a830cd6cc1a9a131f62619","avatarUrl":"/avatars/0c6ba301a66f2db73049c9fe0e97f2ef.svg","isPro":false,"fullname":"Isabelle Mohr","user":"isacat","type":"user"},{"_id":"663ccbff3a74a20189d4aa2e","avatarUrl":"/avatars/83a54455e0157480f65c498cd9057cf2.svg","isPro":false,"fullname":"Nguyen Van Thanh","user":"NguyenVanThanhHust","type":"user"},{"_id":"63107b18e87051f3e3e0f598","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/63107b18e87051f3e3e0f598/R9onir4Y0MZuq1jEWCZ2-.jpeg","isPro":false,"fullname":"Unchun Yang","user":"ucyang","type":"user"}],"acceptLanguages":["*"],"dailyPaperRank":0,"organization":{"_id":"63563e0c2d14fcd7d83743cf","name":"jinaai","fullname":"Jina AI","avatar":"https://cdn-uploads.huggingface.co/production/uploads/603763514de52ff951d89793/wD54VbAHHyHop3uYlJKl4.png"}}">

Papers

arxiv:2408.16672

Jina-ColBERT-v2: A General-Purpose Multilingual Late Interaction Retriever

Published on Aug 29, 2024

· Submitted by

JB D. on Sep 2, 2024

Jina AI

Upvote

Authors:

Bo Wang ,

Michael Günther ,

Saba Sturua ,

Mohammad Kalim Akram ,

Han Xiao

Abstract

Jina-ColBERT-v2 enhances ColBERT's performance and reduces storage requirements through improvements to its architecture and training pipeline, benefiting multilingual retrieval tasks.

AI-generated summary

Multi-vector dense models, such as ColBERT, have proven highly effective in information retrieval. ColBERT's late interaction scoring approximates the joint query-document attention seen in cross-encoders while maintaining inference efficiency closer to traditional dense retrieval models, thanks to its bi-encoder architecture and recent optimizations in indexing and search. In this paper, we introduce several improvements to the ColBERT model architecture and training pipeline, leveraging techniques successful in the more established single-vector embedding model paradigm, particularly those suited for heterogeneous multilingual data. Our new model, Jina-ColBERT-v2, demonstrates strong performance across a range of English and multilingual retrieval tasks, while also cutting storage requirements by up to 50% compared to previous models.