|
--- |
|
license: mit |
|
language: uk |
|
--- |
|
|
|
# roberta-base-wechsel-ukrainian |
|
|
|
[`roberta-base`](https://huggingface.co/roberta-base) transferred to Ukrainian using the method from the NAACL2022 paper [WECHSEL: Effective initialization of subword embeddings for cross-lingual transfer of monolingual language models](https://aclanthology.org/2022.naacl-main.293/). |
|
|
|
# Evaluation |
|
|
|
Evaluation was done on [lang-uk's ner-uk project](https://github.com/lang-uk/ner-uk), the Ukrainian portion of [WikiANN](https://huggingface.co/datasets/wikiann) and the [Ukrainian IU corpus from the Universal Dependencies project](https://github.com/UniversalDependencies/UD_Ukrainian-IU). Evaluation results are the mean of 5 runs with different seeds. |
|
|
|
__Validation Results__ |
|
|
|
| | lang-uk NER (Micro F1) | WikiANN (Micro F1) | UD Ukrainian IU POS (Accuracy) | |
|
|:-------------------------------------------------|:-------------------------|:-------------|:-------------------------| |
|
| roberta-base-wechsel-ukrainian | 88.06 (0.50) | 92.96 (0.08) | 98.70 (0.05) | |
|
| roberta-large-wechsel-ukrainian | __89.27 (0.53)__ | __93.22 (0.15)__ | __98.86 (0.03)__ | |
|
| |
|
| roberta-base-scratch-ukrainian* | 85.49 (0.88) | 91.91 (0.08) | 98.49 (0.04) | |
|
| roberta-large-scratch-ukrainian* | 86.54 (0.70) | 92.39 (0.16) | 98.65 (0.09) | |
|
| |
|
| dbmdz/electra-base-ukrainian-cased-discriminator | 87.49 (0.52) | 93.20 (0.16) | 98.60 (0.03) | |
|
| xlm-roberta-base | 86.68 (0.44) | 92.41 (0.13) | 98.53 (0.02) | |
|
| xlm-roberta-large | 86.64 (1.61) | 93.01 (0.13) | 98.71 (0.04) | |
|
|
|
__Test Results__ |
|
|
|
| | lang-uk NER (Micro F1) | WikiANN (Micro F1) | UD Ukrainian IU POS (Accuracy) | |
|
|:-------------------------------------------------|:-------------------------|:-------------|:-------------------------| |
|
| roberta-base-wechsel-ukrainian | 90.81 (1.51) | 92.98 (0.12) | 98.57 (0.03) | |
|
| roberta-large-wechsel-ukrainian | __91.24 (1.16)__ | __93.22 (0.17)__ | __98.74 (0.06)__ | |
|
| |
|
| roberta-base-scratch-ukrainian* | 89.57 (1.01) | 92.05 (0.09) | 98.31 (0.08) | |
|
| roberta-large-scratch-ukrainian* | 89.96 (0.89) | 92.49 (0.15) | 98.52 (0.04) | |
|
| |
|
| dbmdz/electra-base-ukrainian-cased-discriminator | 90.43 (1.29) | 92.99 (0.11) | 98.59 (0.06) | |
|
| xlm-roberta-base | 90.86 (0.81) | 92.27 (0.09) | 98.45 (0.07) | |
|
| xlm-roberta-large | 90.16 (2.98) | 92.92 (0.19) | 98.71 (0.04) | |
|
|
|
\*trained using the same exact training setup as the wechsel-\* models, but without parameter transfer from WECHSEL. |
|
|
|
|
|
# License |
|
|
|
MIT |
