Matches in SemOpenAlex for { <https://semopenalex.org/work/W4225644300> ?p ?o ?g. }
Showing items 1 to 70 of
70
with 100 items per page.
- W4225644300 abstract "Information retrieval is an important component in natural language processing, for knowledge intensive tasks such as question answering and fact checking. Recently, information retrieval has seen the emergence of dense retrievers, based on neural networks, as an alternative to classical sparse methods based on term-frequency. These models have obtained state-of-the-art results on datasets and tasks where large training sets are available. However, they do not transfer well to new domains or applications with no training data, and are often outperformed by term-frequency methods such as BM25 which are not supervised. Thus, a natural question is whether it is possible to train dense retrievers without supervision. In this work, we explore the limits of contrastive learning as a way to train unsupervised dense retrievers, and show that it leads to strong retrieval performance. More precisely, we show on the BEIR benchmark that our model outperforms BM25 on 11 out of 15 datasets. Furthermore, when a few thousands examples are available, we show that fine-tuning our model on these leads to strong improvements compared to BM25. Finally, when used as pre-training before fine-tuning on the MS-MARCO dataset, our technique obtains state-of-the-art results on the BEIR benchmark." @default.
- W4225644300 created "2022-05-05" @default.
- W4225644300 creator A5001151643 @default.
- W4225644300 creator A5007815326 @default.
- W4225644300 creator A5035420035 @default.
- W4225644300 creator A5041693326 @default.
- W4225644300 creator A5069316249 @default.
- W4225644300 creator A5070305512 @default.
- W4225644300 creator A5071626315 @default.
- W4225644300 date "2021-12-16" @default.
- W4225644300 modified "2023-09-24" @default.
- W4225644300 title "Towards Unsupervised Dense Information Retrieval with Contrastive Learning" @default.
- W4225644300 hasPublicationYear "2021" @default.
- W4225644300 type Work @default.
- W4225644300 citedByCount "0" @default.
- W4225644300 crossrefType "posted-content" @default.
- W4225644300 hasAuthorship W4225644300A5001151643 @default.
- W4225644300 hasAuthorship W4225644300A5007815326 @default.
- W4225644300 hasAuthorship W4225644300A5035420035 @default.
- W4225644300 hasAuthorship W4225644300A5041693326 @default.
- W4225644300 hasAuthorship W4225644300A5069316249 @default.
- W4225644300 hasAuthorship W4225644300A5070305512 @default.
- W4225644300 hasAuthorship W4225644300A5071626315 @default.
- W4225644300 hasBestOaLocation W42256443001 @default.
- W4225644300 hasConcept C119857082 @default.
- W4225644300 hasConcept C121332964 @default.
- W4225644300 hasConcept C13280743 @default.
- W4225644300 hasConcept C150899416 @default.
- W4225644300 hasConcept C154945302 @default.
- W4225644300 hasConcept C168167062 @default.
- W4225644300 hasConcept C185798385 @default.
- W4225644300 hasConcept C204321447 @default.
- W4225644300 hasConcept C205649164 @default.
- W4225644300 hasConcept C23123220 @default.
- W4225644300 hasConcept C41008148 @default.
- W4225644300 hasConcept C44291984 @default.
- W4225644300 hasConcept C61797465 @default.
- W4225644300 hasConcept C62520636 @default.
- W4225644300 hasConcept C97355855 @default.
- W4225644300 hasConceptScore W4225644300C119857082 @default.
- W4225644300 hasConceptScore W4225644300C121332964 @default.
- W4225644300 hasConceptScore W4225644300C13280743 @default.
- W4225644300 hasConceptScore W4225644300C150899416 @default.
- W4225644300 hasConceptScore W4225644300C154945302 @default.
- W4225644300 hasConceptScore W4225644300C168167062 @default.
- W4225644300 hasConceptScore W4225644300C185798385 @default.
- W4225644300 hasConceptScore W4225644300C204321447 @default.
- W4225644300 hasConceptScore W4225644300C205649164 @default.
- W4225644300 hasConceptScore W4225644300C23123220 @default.
- W4225644300 hasConceptScore W4225644300C41008148 @default.
- W4225644300 hasConceptScore W4225644300C44291984 @default.
- W4225644300 hasConceptScore W4225644300C61797465 @default.
- W4225644300 hasConceptScore W4225644300C62520636 @default.
- W4225644300 hasConceptScore W4225644300C97355855 @default.
- W4225644300 hasLocation W42256443001 @default.
- W4225644300 hasOpenAccess W4225644300 @default.
- W4225644300 hasPrimaryLocation W42256443001 @default.
- W4225644300 hasRelatedWork W10714513 @default.
- W4225644300 hasRelatedWork W13972091 @default.
- W4225644300 hasRelatedWork W17086583 @default.
- W4225644300 hasRelatedWork W2060686 @default.
- W4225644300 hasRelatedWork W4281541 @default.
- W4225644300 hasRelatedWork W5460003 @default.
- W4225644300 hasRelatedWork W5772686 @default.
- W4225644300 hasRelatedWork W842930 @default.
- W4225644300 hasRelatedWork W8965072 @default.
- W4225644300 hasRelatedWork W9662544 @default.
- W4225644300 isParatext "false" @default.
- W4225644300 isRetracted "false" @default.
- W4225644300 workType "article" @default.