Jina-ColBERT-v2: A General-Purpose Multilingual Late Interaction Retriever
Fuente:
arXiv
Salvato in:
| Autori principali: | , , , , , , , , , |
|---|---|
| Natura: | Preprint |
| Pubblicazione: |
2024
|
| Soggetti: | |
| Accesso online: | |
| Tags: |
Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
|
| _version_ | 1866909316046389248 |
|---|---|
| author | Jha, Rohan Wang, Bo Günther, Michael Mastrapas, Georgios Sturua, Saba Mohr, Isabelle Koukounas, Andreas Akram, Mohammad Kalim Wang, Nan Xiao, Han |
| author_facet | Jha, Rohan Wang, Bo Günther, Michael Mastrapas, Georgios Sturua, Saba Mohr, Isabelle Koukounas, Andreas Akram, Mohammad Kalim Wang, Nan Xiao, Han |
| contents | 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 work we propose a number of incremental improvements to the ColBERT model architecture and training pipeline, using methods shown to work in the more mature single-vector embedding model training paradigm, particularly those that apply to heterogeneous multilingual data or boost efficiency with little tradeoff. Our new model, Jina-ColBERT-v2, demonstrates strong performance across a range of English and multilingual retrieval tasks. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2408_16672 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | Jina-ColBERT-v2: A General-Purpose Multilingual Late Interaction Retriever Jha, Rohan Wang, Bo Günther, Michael Mastrapas, Georgios Sturua, Saba Mohr, Isabelle Koukounas, Andreas Akram, Mohammad Kalim Wang, Nan Xiao, Han Information Retrieval Artificial Intelligence Computation and Language 68T50 I.2.7 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 work we propose a number of incremental improvements to the ColBERT model architecture and training pipeline, using methods shown to work in the more mature single-vector embedding model training paradigm, particularly those that apply to heterogeneous multilingual data or boost efficiency with little tradeoff. Our new model, Jina-ColBERT-v2, demonstrates strong performance across a range of English and multilingual retrieval tasks. |
| title | Jina-ColBERT-v2: A General-Purpose Multilingual Late Interaction Retriever |
| topic | Information Retrieval Artificial Intelligence Computation and Language 68T50 I.2.7 |
| url | https://arxiv.org/abs/2408.16672 |