File size: 17,920 Bytes
40726ee aba5939 40726ee aba5939 40726ee aba5939 40726ee 9a39a0d 40726ee aba5939 40726ee aba5939 40726ee aba5939 40726ee aba5939 40726ee aba5939 40726ee aba5939 40726ee aba5939 40726ee aba5939 40726ee aba5939 40726ee aba5939 40726ee 9a39a0d 40726ee aba5939 40726ee aba5939 40726ee aba5939 40726ee aba5939 6aa2298 aba5939 40726ee 6aa2298 40726ee aba5939 40726ee aba5939 40726ee aba5939 40726ee aba5939 40726ee aba5939 40726ee aba5939 40726ee aba5939 40726ee aba5939 40726ee aba5939 |
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 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 |
---
license: cc-by-4.0
language:
- cs
- pl
- sk
- sl
- en
library_name: transformers
tags:
- translation
- mt
- marian
- pytorch
- sentence-piece
- multilingual
- allegro
- laniqo
---
# MultiSlav BiDi Models
<p align="center">
<a href="https://ml.allegro.tech/"><img src="allegro-title.svg" alt="MLR @ Allegro.com"></a>
</p>
## Multilingual BiDi MT Models
___BiDi___ is a collection of Encoder-Decoder vanilla transformer models trained on sentence-level Machine Translation task.
Each model is supporting Bi-Directional translation.
___BiDi___ models are part of the [___MultiSlav___ collection](https://huggingface.co/collections/allegro/multislav-6793d6b6419e5963e759a683). More information will be available soon in our upcoming MultiSlav paper.
Experiments were conducted under research project by [Machine Learning Research](https://ml.allegro.tech/) lab for [Allegro.com](https://ml.allegro.tech/).
Big thanks to [laniqo.com](laniqo.com) for cooperation in the research.
<p align="center">
<img src="bi-di.svg">
</p>
Graphic above provides an example of an BiDi model - [BiDi-ces-pol](https://huggingface.co/allegro/bidi-ces-pol) to translate from Polish to Czech language.
___BiDi-ces-pol___ is a bi-directional model supporting translation both __form Czech to Polish__ and __from Polish to Czech__ directions.
### Supported languages
To use a ___BiDi___ model, you must provide the target language for translation.
Target language tokens are represented as 3-letter ISO 639-3 language codes embedded in a format >>xxx<<.
All accepted directions and their respective tokens are listed below.
Note that, for each model only two directions are available.
Each of them was added as a special token to Sentence-Piece tokenizer.
| **Target Language** | **First token** |
|---------------------|-----------------|
| Czech | `>>ces<<` |
| English | `>>eng<<` |
| Polish | `>>pol<<` |
| Slovak | `>>slk<<` |
| Slovene | `>>slv<<` |
### Bi-Di models available
We provided 10 ___BiDi___ models, allowing to translate between 20 languages.
| **Bi-Di model** | **Languages supported** | **HF repository** |
|-----------------|-------------------------|---------------------------------------------------------------------|
| BiDi-ces-eng | Czech ↔ English | [allegro/BiDi-ces-eng](https://huggingface.co/allegro/bidi-ces-eng) |
| BiDi-ces-pol | Czech ↔ Polish | [allegro/BiDi-ces-pol](https://huggingface.co/allegro/bidi-ces-pol) |
| BiDi-ces-slk | Czech ↔ Slovak | [allegro/BiDi-ces-slk](https://huggingface.co/allegro/bidi-ces-slk) |
| BiDi-ces-slv | Czech ↔ Slovene | [allegro/BiDi-ces-slv](https://huggingface.co/allegro/bidi-ces-slv) |
| BiDi-eng-pol | English ↔ Polish | [allegro/BiDi-eng-pol](https://huggingface.co/allegro/bidi-eng-pol) |
| BiDi-eng-slk | English ↔ Slovak | [allegro/BiDi-eng-slk](https://huggingface.co/allegro/bidi-eng-slk) |
| BiDi-eng-slv | English ↔ Slovene | [allegro/BiDi-eng-slv](https://huggingface.co/allegro/bidi-eng-slv) |
| BiDi-pol-slk | Polish ↔ Slovak | [allegro/BiDi-pol-slk](https://huggingface.co/allegro/bidi-pol-slk) |
| BiDi-pol-slv | Polish ↔ Slovene | [allegro/BiDi-pol-slv](https://huggingface.co/allegro/bidi-pol-slv) |
| BiDi-slk-slv | Slovak ↔ Slovene | [allegro/BiDi-slk-slv](https://huggingface.co/allegro/bidi-slk-slv) |
## Use case quickstart
Example code-snippet to use model. Due to bug the `MarianMTModel` must be used explicitly.
Remember to adjust source and target languages to your use-case.
```python
from transformers import AutoTokenizer, MarianMTModel
source_lang = "pol"
target_lang = "ces"
first_lang, second_lang = sorted([source_lang, target_lang])
model_name = f"Allegro/BiDi-{first_lang}-{second_lang}"
tokenizer = AutoTokenizer.from_pretrained(model_name)
model = MarianMTModel.from_pretrained(model_name)
text = f">>{target_lang}<<" + " " + "Allegro to internetowa platforma e-commerce, na której swoje produkty sprzedają średnie i małe firmy, jak również duże marki."
batch_to_translate = [text]
translations = model.generate(**tokenizer.batch_encode_plus(batch_to_translate, return_tensors="pt"))
decoded_translation = tokenizer.batch_decode(translations, skip_special_tokens=True, clean_up_tokenization_spaces=True)[0]
print(decoded_translation)
```
Generated Czech output:
> Allegro je online e-commerce platforma, na které své výrobky prodávají střední a malé firmy, stejně jako velké značky.
## Training
[SentencePiece](https://github.com/google/sentencepiece) tokenizer has a vocab size 32k in total (16k per language). Tokenizer was trained on randomly sampled part of the training corpus.
During the training we used the [MarianNMT](https://marian-nmt.github.io/) framework.
Base marian configuration used: [transfromer-big](https://github.com/marian-nmt/marian-dev/blob/master/src/common/aliases.cpp#L113).
All training parameters are listed in table below.
### Training hyperparameters:
| **Hyperparameter** | **Value** |
|----------------------------|------------------------------------------------------------------------------------------------------------|
| Total Parameter Size | 209M |
| Vocab Size | 32k |
| Base Parameters | [Marian transfromer-big](https://github.com/marian-nmt/marian-dev/blob/master/src/common/aliases.cpp#L113) |
| Number of Encoding Layers | 6 |
| Number of Decoding Layers | 6 |
| Model Dimension | 1024 |
| FF Dimension | 4096 |
| Heads | 16 |
| Dropout | 0.1 |
| Batch Size | mini batch fit to VRAM |
| Training Accelerators | 4x A100 40GB |
| Max Length | 100 tokens |
| Optimizer | Adam |
| Warmup steps | 8000 |
| Context | Sentence-level MT |
| Languages Supported | See [Bi-Di models available](#Bi-Di-models-available) |
| Precision | float16 |
| Validation Freq | 3000 steps |
| Stop Metric | ChrF |
| Stop Criterion | 20 Validation steps |
## Training corpora
The main research question was: "How does adding additional, related languages impact the quality of the model?" - we explored it in the Slavic language family.
___BiDi___ models are our baseline before expanding the data-regime by using higher-level multilinguality.
Datasets were downloaded via [MT-Data](https://pypi.org/project/mtdata/0.2.10/) library.
The number of total examples post filtering and deduplication varies, depending on languages supported, see the table below.
| **Language pair** | **Number of training examples** |
|-------------------|--------------------------------:|
| Czech ↔ Polish | 63M |
| Czech ↔ Slovak | 30M |
| Czech ↔ Slovene | 25M |
| Polish ↔ Slovak | 26M |
| Polish ↔ Slovene | 23M |
| Slovak ↔ Slovene | 18M |
| ---------------- | ------------------------------- |
| Czech ↔ English | 151M |
| English ↔ Polish | 150M |
| English ↔ Slovak | 52M |
| English ↔ Slovene | 40M |
The datasets used (only applicable to specific directions):
| **Corpus** |
|----------------------|
| paracrawl |
| opensubtitles |
| multiparacrawl |
| dgt |
| elrc |
| xlent |
| wikititles |
| wmt |
| wikimatrix |
| dcep |
| ELRC |
| tildemodel |
| europarl |
| eesc |
| eubookshop |
| emea |
| jrc_acquis |
| ema |
| qed |
| elitr_eca |
| EU-dcep |
| rapid |
| ecb |
| kde4 |
| news_commentary |
| kde |
| bible_uedin |
| europat |
| elra |
| wikipedia |
| wikimedia |
| tatoeba |
| globalvoices |
| euconst |
| ubuntu |
| php |
| ecdc |
| eac |
| eac_reference |
| gnome |
| EU-eac |
| books |
| EU-ecdc |
| newsdev |
| khresmoi_summary |
| czechtourism |
| khresmoi_summary_dev |
| worldbank |
## Evaluation
Evaluation of the models was performed on [Flores200](https://huggingface.co/datasets/facebook/flores) dataset.
The table below compares performance of the open-source models and all applicable models from our collection.
Metric used: Unbabel/wmt22-comet-da.
| **Direction** | **CES → ENG** | **CES → POL** | **CES → SLK** | **CES → SLV** | **ENG → CES** | **ENG → POL** | **ENG → SLK** | **ENG → SLV** | **POL → CES** | **POL → ENG** | **POL → SLK** | **POL → SLV** | **SLK → CES** | **SLK → ENG** | **SLK → POL** | **SLK → SLV** | **SLV → CES** | **SLV → ENG** | **SLV → POL** | **SLV → SLK** |
|----------------------------------------------------|--------------:|--------------:|--------------:|--------------:|--------------:|--------------:|--------------:|--------------:|--------------:|--------------:|--------------:|--------------:|--------------:|--------------:|--------------:|--------------:|--------------:|--------------:|--------------:|--------------:|
| **M2M-100** | 87.0 | 89.0 | 92.1 | 89.7 | 88.6 | 86.4 | 88.4 | 87.3 | 89.6 | 84.6 | 89.4 | 88.4 | 92.7 | 86.8 | 89.1 | 89.6 | 90.3 | 86.4 | 88.7 | 90.1 |
| **NLLB-200** | 88.1 | 88.9 | 91.2 | 88.6 | 90.4 | __88.5__ | 90.1 | 88.8 | 89.4 | __85.8__ | 88.9 | 87.7 | 91.8 | 88.2 | 88.9 | 88.8 | 90.0 | __87.5__ | 88.6 | 89.4 |
| **Seamless-M4T** | 87.5 | 80.9 | 90.8 | 82.0 | __90.7__ | __88.5__ | __90.6__ | __89.6__ | 79.6 | 85.4 | 80.0 | 76.4 | 91.5 | 87.2 | 81.2 | 82.9 | 80.9 | 87.3 | 76.7 | 81.0 |
| **OPUS-MT Sla-Sla** | __88.2__ | 82.8 | - | 83.4 | 89.1 | 85.6 | - | 84.5 | 82.9 | 82.2 | - | 81.2 | - | - | - | - | 83.5 | 84.1 | 80.8 | - |
| **OPUS-MT SK-EN** | - | - | - | - | - | - | 89.5 | - | - | - | - | - | - | __88.4__ | - | - | - | - | - | - |
| _Our contributions:_ | | | | | | | | | | | | | | | | | | | | |
| **BiDi Models**<span style="color:green;">*</span> | 87.5 | 89.4 | 92.4 | 89.8 | 87.8 | 86.2 | 87.2 | 86.6 | 90.0 | 85.0 | 89.1 | 88.4 | 92.9 | 87.3 | 88.8 | 89.4 | 90.0 | 86.9 | 88.1 | 89.1 |
| **P4-pol**<span style="color:red;">◊</span> | - | 89.6 | 90.8 | 88.7 | - | - | - | - | 90.2 | - | 89.8 | 88.7 | 91.0 | - | 89.3 | 88.4 | 89.3 | - | 88.7 | 88.5 |
| **P5-eng**<span style="color:red;">◊</span> | 88.0 | 89.0 | 90.7 | 89.0 | 88.8 | 87.3 | 88.4 | 87.5 | 89.0 | 85.7 | 88.5 | 87.8 | 91.0 | 88.2 | 88.6 | 88.5 | 89.6 | 87.2 | 88.4 | 88.9 |
| **P5-ces**<span style="color:red;">◊</span> | 87.9 | 89.6 | __92.5__ | 89.9 | 88.4 | 85.0 | 87.9 | 85.9 | 90.3 | 84.5 | 89.5 | 88.0 | __93.0__ | 87.8 | 89.4 | 89.8 | 90.3 | 85.7 | 87.9 | 89.8 |
| **MultiSlav-4slav** | - | 89.7 | __92.5__ | 90.0 | - | - | - | - | 90.2 | - | 89.6 | 88.7 | 92.9 | - | 89.4 | 90.1 | __90.6__ | - | 88.9 | __90.2__ |
| **MultiSlav-5lang** | 87.8 | __89.8__ | __92.5__ | __90.1__ | 88.9 | 86.9 | 88.0 | 87.3 | __90.4__ | 85.4 | 89.8 | __88.9__ | 92.9 | 87.8 | __89.6__ | __90.2__ | __90.6__ | 87.0 | __89.2__ | __90.2__ |
<span style="color:red;">◊</span> system of 2 models *Many2XXX* and *XXX2Many*, see [P5-ces2many](https://huggingface.co/allegro/p5-ces2many)
<span style="color:green;">*</span> results combined for all bi-directional models; each values for applicable model
## Limitations and Biases
We did not evaluate inherent bias contained in training datasets. It is advised to validate bias of our models in perspective domain. This might be especially problematic in translation from English to Slavic languages, which require explicitly indicated gender and might hallucinate based on bias present in training data.
## License
The model is licensed under CC BY 4.0, which allows for commercial use.
## Citation
TO BE UPDATED SOON 🤗
## Contact Options
Authors:
- MLR @ Allegro: [Artur Kot](https://linkedin.com/in/arturkot), [Mikołaj Koszowski](https://linkedin.com/in/mkoszowski), [Wojciech Chojnowski](https://linkedin.com/in/wojciech-chojnowski-744702348), [Mieszko Rutkowski](https://linkedin.com/in/mieszko-rutkowski)
- Laniqo.com: [Artur Nowakowski](https://linkedin.com/in/artur-nowakowski-mt), [Kamil Guttmann](https://linkedin.com/in/kamil-guttmann), [Mikołaj Pokrywka](https://linkedin.com/in/mikolaj-pokrywka)
Please don't hesitate to contact authors if you have any questions or suggestions:
- e-mail: [email protected] or [email protected]
- LinkedIn: [Artur Kot](https://linkedin.com/in/arturkot) or [Mikołaj Koszowski](https://linkedin.com/in/mkoszowski)
|