| 1. |
|
|
| 2. |
|
|
| 3. |
|
|
| 4. |
|
|
| 5. |
- Abbafati, Cristiana, et al.
(författare)
-
- 2020
-
Tidskriftsartikel (refereegranskat)
|
|
| 6. |
- Adelani, David, et al.
(författare)
-
A Few Thousand Translations Go A Long Way! Leveraging Pre-trained Models for African News Translation
- 2022
-
Ingår i: NAACL 2022. - Stroudsburg : Association for Computational Linguistics. - 9781955917711 ; , s. 3053-3070
-
Konferensbidrag (refereegranskat)abstract
- Recent advances in the pre-training of language models leverage large-scale datasets to create multilingual models. However, low-resource languages are mostly left out in these datasets. This is primarily because many widely spoken languages are not well represented on the web and therefore excluded from the large-scale crawls used to create datasets. Furthermore, downstream users of these models are restricted to the selection of languages originally chosen for pre-training. This work investigates how to optimally leverage existing pre-trained models to create low-resource translation systems for 16 African languages. We focus on two questions: 1) How can pre-trained models be used for languages not included in the initial pre-training? and 2) How can the resulting translation models effectively transfer to new domains? To answer these questions, we create a new African news corpus covering 16 languages, of which eight languages are not part of any existing evaluation dataset. We demonstrate that the most effective strategy for transferring both to additional languages and to additional domains is to fine-tune large pre-trained models on small quantities of high-quality translation data.
|
|
| 7. |
- Adelani, David Ifeoluwa, et al.
(författare)
-
MasakhaNER 2.0: Africa-centric Transfer Learning for Named Entity Recognition
- 2022
-
Ingår i: Proceedings of the 2022 Conference on Empirical Methods in Natural Language Processing. - : Association for Computational Linguistics (ACL). ; , s. 4488-4508
-
Konferensbidrag (refereegranskat)abstract
- African languages are spoken by over a billion people, but are underrepresented in NLP research and development. The challenges impeding progress include the limited availability of annotated datasets, as well as a lack of understanding of the settings where current methods are effective. In this paper, we make progress towards solutions for these challenges, focusing on the task of named entity recognition (NER). We create the largest human-annotated NER dataset for 20 African languages, and we study the behavior of state-of-the-art cross-lingual transfer methods in an Africa-centric setting, demonstrating that the choice of source language significantly affects performance. We show that choosing the best transfer language improves zero-shot F1 scores by an average of 14 points across 20 languages compared to using English. Our results highlight the need for benchmark datasets and models that cover typologically-diverse African languages.
|
|
| 8. |
- Haridass, Isha N., et al.
(författare)
-
Cellular metabolism and pore lifetime of human skin following microprojection array mediation
- 2019
-
Ingår i: Journal of Controlled Release. - : ELSEVIER SCIENCE BV. - 0168-3659 .- 1873-4995. ; 306, s. 59-68
-
Tidskriftsartikel (refereegranskat)abstract
- Skin-targeting microscale medical devices are becoming popular for therapeutic delivery and diagnosis. We used cryo-SEM, fluorescence lifetime imaging microscopy (FLIM), autofluorescence imaging microscopy and inflammatory response to study the puncturing and recovery of human skin ex vivo and in vivo after discretised puncturing by a microneedle array (Nanopatch (R)). Pores induced by the microprojections were found to close by similar to 25% in diameter within the first 30 min, and almost completely close by similar to 6 h. FLIM images of ex vivo viable epidermis showed a stable fluorescence lifetime for unpatched areas of similar to 1000 ps up to 24 h. Only the cells in the immediate puncture zones (in direct contact with projections) showed a reduction in the observed fluorescence lifetimes to between similar to 518-583 ps. The ratio of free-bound NAD(P)H (alpha 1/alpha 2) in unaffected areas of the viable epidermis was similar to 2.5-3.0, whereas the ratio at puncture holes was almost double at similar to 4.2-4.6. An exploratory pilot in vivo study also suggested similar closure rate with histamine administration to the forearms of human volunteers after Nanopatch (R) treatment, although a prolonged inflammation was observed with Tissue Viability Imaging. Overall, this work shows that the pores created by the microneedle-type medical device, Nanopatch (R), are transient, with the skin recovering rapidly within 1-2 days in the epidermis after application.
|
|
| 9. |
|
|
| 10. |
|
|
