---
language:
- nl
base_model:
- FacebookAI/xlm-roberta-base
pipeline_tag: token-classification
tags:
- coreference
- resolution
- gender
- neutral
- pronouns
- debiading
- dutch
- CDA
---
This Dutch coreference resolution model is based on the [wl-coref](https://github.com/vdobrovolskii/wl-coref) model. The model was debiased through CDA, in order to improve its performance on gender-neutral pronouns and neopronouns. We used [XLM-RoBERTa-base](https://huggingface.co/FacebookAI/xlm-roberta-base) as our base model, fine-tuned the model on the [SoNaR-1 corpus](https://taalmaterialen.ivdnt.org/download/tstc-sonar-corpus/), and then further fine-tuned the model on a gender-neutral version of the this corpus for debiasing purposes. We used five different seeds during our experiment, and upload all five versions of the model.

For more information, see our [Github repository](https://github.com/gvanboven/Transforming_Dutch). 

This model was created as part of our FAccT 2024 paper. Find our published paper at https://dl.acm.org/doi/10.1145/3630106.3659049 and arxiv paper at https://arxiv.org/abs/2405.00134. 

## Usage
For usage instructions for the wl-coref model, see [their repo](https://github.com/vdobrovolskii/wl-coref). 

## Citation
G. van Boven, Y. Du, D. Nguyen, _Transforming Dutch: Debiasing Dutch Coreference Resolution Systems for Non-binary Pronouns_. FAccT 2024.

```
@inproceedings{boven-2024-transforming,
title = "Transforming Dutch: Debiasing Dutch Coreference Resolution Systems for Non-binary Pronouns",
author = "van Boven, Goya and Du, Yupei and Nguyen, Dong",
booktitle = "Proceedings of the 2024 Conference on Fairness, Accountability, and Transparency",
month = jun,
year = "2024",
address = "Online and Rio de Janeiro, Brazil",
publisher = "Association for Computing Machinery"
}
```