Matches in SemOpenAlex for { <https://semopenalex.org/work/W4385571209> ?p ?o ?g. }
Showing items 1 to 77 of
77
with 100 items per page.
- W4385571209 abstract "Most existing word alignment methods rely on manual alignment datasets or parallel corpora, which limits their usefulness. Here, to mitigate the dependence on manual data, we broaden the source of supervision by relaxing the requirement for correct, fully-aligned, and parallel sentences. Specifically, we make noisy, partially aligned, and non-parallel paragraphs in this paper. We then use such a large-scale weakly-supervised dataset for word alignment pre-training via span prediction. Extensive experiments with various settings empirically demonstrate that our approach, which is named WSPAlign, is an effective and scalable way to pre-train word aligners without manual data. When fine-tuned on standard benchmarks, WSPAlign has set a new state of the art by improving upon the best supervised baseline by 3.3 6.1 points in F1 and 1.5 6.1 points in AER. Furthermore, WSPAlign also achieves competitive performance compared with the corresponding baselines in few-shot, zero-shot and cross-lingual tests, which demonstrates that WSPAlign is potentially more practical for low-resource languages than existing methods." @default.
- W4385571209 created "2023-08-05" @default.
- W4385571209 creator A5057313515 @default.
- W4385571209 creator A5064113904 @default.
- W4385571209 creator A5077525450 @default.
- W4385571209 date "2023-01-01" @default.
- W4385571209 modified "2023-09-24" @default.
- W4385571209 title "WSPAlign: Word Alignment Pre-training via Large-Scale Weakly Supervised Span Prediction" @default.
- W4385571209 doi "https://doi.org/10.18653/v1/2023.acl-long.621" @default.
- W4385571209 hasPublicationYear "2023" @default.
- W4385571209 type Work @default.
- W4385571209 citedByCount "0" @default.
- W4385571209 crossrefType "proceedings-article" @default.
- W4385571209 hasAuthorship W4385571209A5057313515 @default.
- W4385571209 hasAuthorship W4385571209A5064113904 @default.
- W4385571209 hasAuthorship W4385571209A5077525450 @default.
- W4385571209 hasBestOaLocation W43855712091 @default.
- W4385571209 hasConcept C111368507 @default.
- W4385571209 hasConcept C119857082 @default.
- W4385571209 hasConcept C121332964 @default.
- W4385571209 hasConcept C12725497 @default.
- W4385571209 hasConcept C127313418 @default.
- W4385571209 hasConcept C127413603 @default.
- W4385571209 hasConcept C147176958 @default.
- W4385571209 hasConcept C153180895 @default.
- W4385571209 hasConcept C154945302 @default.
- W4385571209 hasConcept C177264268 @default.
- W4385571209 hasConcept C199360897 @default.
- W4385571209 hasConcept C204321447 @default.
- W4385571209 hasConcept C2524010 @default.
- W4385571209 hasConcept C2778753569 @default.
- W4385571209 hasConcept C2778755073 @default.
- W4385571209 hasConcept C33923547 @default.
- W4385571209 hasConcept C41008148 @default.
- W4385571209 hasConcept C48044578 @default.
- W4385571209 hasConcept C51632099 @default.
- W4385571209 hasConcept C62520636 @default.
- W4385571209 hasConcept C77088390 @default.
- W4385571209 hasConcept C90805587 @default.
- W4385571209 hasConceptScore W4385571209C111368507 @default.
- W4385571209 hasConceptScore W4385571209C119857082 @default.
- W4385571209 hasConceptScore W4385571209C121332964 @default.
- W4385571209 hasConceptScore W4385571209C12725497 @default.
- W4385571209 hasConceptScore W4385571209C127313418 @default.
- W4385571209 hasConceptScore W4385571209C127413603 @default.
- W4385571209 hasConceptScore W4385571209C147176958 @default.
- W4385571209 hasConceptScore W4385571209C153180895 @default.
- W4385571209 hasConceptScore W4385571209C154945302 @default.
- W4385571209 hasConceptScore W4385571209C177264268 @default.
- W4385571209 hasConceptScore W4385571209C199360897 @default.
- W4385571209 hasConceptScore W4385571209C204321447 @default.
- W4385571209 hasConceptScore W4385571209C2524010 @default.
- W4385571209 hasConceptScore W4385571209C2778753569 @default.
- W4385571209 hasConceptScore W4385571209C2778755073 @default.
- W4385571209 hasConceptScore W4385571209C33923547 @default.
- W4385571209 hasConceptScore W4385571209C41008148 @default.
- W4385571209 hasConceptScore W4385571209C48044578 @default.
- W4385571209 hasConceptScore W4385571209C51632099 @default.
- W4385571209 hasConceptScore W4385571209C62520636 @default.
- W4385571209 hasConceptScore W4385571209C77088390 @default.
- W4385571209 hasConceptScore W4385571209C90805587 @default.
- W4385571209 hasLocation W43855712091 @default.
- W4385571209 hasOpenAccess W4385571209 @default.
- W4385571209 hasPrimaryLocation W43855712091 @default.
- W4385571209 hasRelatedWork W1525643724 @default.
- W4385571209 hasRelatedWork W2067938758 @default.
- W4385571209 hasRelatedWork W2129217837 @default.
- W4385571209 hasRelatedWork W2364921833 @default.
- W4385571209 hasRelatedWork W2382623646 @default.
- W4385571209 hasRelatedWork W2792951589 @default.
- W4385571209 hasRelatedWork W2952340579 @default.
- W4385571209 hasRelatedWork W2961085424 @default.
- W4385571209 hasRelatedWork W3087771547 @default.
- W4385571209 hasRelatedWork W3201070945 @default.
- W4385571209 isParatext "false" @default.
- W4385571209 isRetracted "false" @default.
- W4385571209 workType "article" @default.