File size: 3,167 Bytes
da6967f
d63f342
 
da6967f
 
592f960
da6967f
a40f97a
da6967f
d63f342
da6967f
3c755b0
da6967f
d63f342
da6967f
d63f342
 
 
 
3c755b0
d63f342
 
3c755b0
d63f342
 
 
da6967f
d63f342
da6967f
d63f342
 
 
 
3c755b0
d63f342
 
3c755b0
d63f342
 
 
da6967f
3c755b0
da6967f
 
d63f342
da6967f
d63f342
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
---
license: mit
language: uk
---

# roberta-large-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