README.md

---
language:
- en
- fr
- it
- pt
tags:
- formality
licenses:
- cc-by-nc-sa
license: openrail++
base_model:
- distilbert/distilbert-base-multilingual-cased
---


**Model Overview**

This is the model presented in the paper ["Detecting Text Formality: A Study of Text Classification Approaches"](https://aclanthology.org/2023.ranlp-1.31/). 

The original model is [mDistilBERT (base)](https://huggingface.co/distilbert-base-multilingual-cased). Then, it was fine-tuned on the multilingual corpus for fomality classiication [X-FORMAL](https://arxiv.org/abs/2104.04108) that consists of 4 languages -- English (from [GYAFC](https://arxiv.org/abs/1803.06535)), French, Italian, and Brazilian Portuguese.
In our experiments, the model showed the best results within Transformer-based models for the cross-lingual formality classification knowledge transfer task. More details, code and data can be found [here](https://github.com/s-nlp/formality).

**Evaluation Results**

Here, we provide several metrics of the best models from each category participated in the comparison to understand the ranks of values. We report accuracy score for two setups -- multilingual model fine-tuned for each language separately and then fine-tuned on all languages.
For cross-lingual experiments results, please, refer to the paper.

|                  | En   |  It  |  Po  |  Fr  |  All  |
|------------------|------|------|------|------|-------|
| bag-of-words     | 79.1 | 71.3 | 70.6 | 72.5 |  ---  |
| CharBiLSTM       | 87.0 | 79.1 | 75.9 | 81.3 |  82.7 |
| mDistilBERT-cased| 86.6 | 76.8 | 75.9 | 79.1 |  79.4 |
| mDeBERTa-base    | 87.3 | 76.6 | 75.8 | 78.9 |  79.9 |

**How to use**

```python
from transformers import AutoModelForSequenceClassification, AutoTokenizer
model_name = 's-nlp/mdistilbert-base-formality-ranker'
tokenizer = AutoTokenizer.from_pretrained(model_name)
model = AutoModelForSequenceClassification.from_pretrained(model_name)
```

**Citation**
```
@inproceedings{dementieva-etal-2023-detecting,
    title = "Detecting Text Formality: A Study of Text Classification Approaches",
    author = "Dementieva, Daryna  and
      Babakov, Nikolay  and
      Panchenko, Alexander",
    editor = "Mitkov, Ruslan  and
      Angelova, Galia",
    booktitle = "Proceedings of the 14th International Conference on Recent Advances in Natural Language Processing",
    month = sep,
    year = "2023",
    address = "Varna, Bulgaria",
    publisher = "INCOMA Ltd., Shoumen, Bulgaria",
    url = "https://aclanthology.org/2023.ranlp-1.31",
    pages = "274--284",
    abstract = "Formality is one of the important characteristics of text documents. The automatic detection of the formality level of a text is potentially beneficial for various natural language processing tasks. Before, two large-scale datasets were introduced for multiple languages featuring formality annotation{---}GYAFC and X-FORMAL. However, they were primarily used for the training of style transfer models. At the same time, the detection of text formality on its own may also be a useful application. This work proposes the first to our knowledge systematic study of formality detection methods based on statistical, neural-based, and Transformer-based machine learning methods and delivers the best-performing models for public usage. We conducted three types of experiments {--} monolingual, multilingual, and cross-lingual. The study shows the overcome of Char BiLSTM model over Transformer-based ones for the monolingual and multilingual formality classification task, while Transformer-based classifiers are more stable to cross-lingual knowledge transfer.",
}
```

## Licensing Information

This model is licensed under the OpenRAIL++ License, which supports the development of various technologies—both industrial and academic—that serve the public good.