File size: 2,102 Bytes
22f63f7
 
59c8b37
 
 
 
22f63f7
59c8b37
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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
---
license: mit
language:
- fro
pipeline_tag: fill-mask
library_name: transformers
---

BERTrade-base
=============

This contextual embeddings model for Old French was developped as part of the work of Grobol, Regnault, Ortiz-Suarez et al. ([2022](https://aclanthology.org/2022.lrec-1.119/)), we refer to it for a list of the resources and the settings used to develop this model as well as assessment of their suitability for syntactic dependency parsing of Old French.

If you use it, please cite the original paper:

```bibtex
@inproceedings{grobol-etal-2022-bertrade,
    title = "{BERT}rade: Using Contextual Embeddings to Parse {O}ld {F}rench",
    author = {Grobol, Lo{\"\i}c  and
      Regnault, Mathilde  and
      Ortiz Suarez, Pedro  and
      Sagot, Beno{\^\i}t  and
      Romary, Laurent  and
      Crabb{\'e}, Benoit},
    booktitle = "Proceedings of the Thirteenth Language Resources and Evaluation Conference",
    month = jun,
    year = "2022",
    address = "Marseille, France",
    publisher = "European Language Resources Association",
    url = "https://aclanthology.org/2022.lrec-1.119",
    pages = "1104--1113",
    abstract = "The successes of contextual word embeddings learned by training large-scale language models, while remarkable, have mostly occurred for languages where significant amounts of raw texts are available and where annotated data in downstream tasks have a relatively regular spelling. Conversely, it is not yet completely clear if these models are also well suited for lesser-resourced and more irregular languages. We study the case of Old French, which is in the interesting position of having relatively limited amount of available raw text, but enough annotated resources to assess the relevance of contextual word embedding models for downstream NLP tasks. In particular, we use POS-tagging and dependency parsing to evaluate the quality of such models in a large array of configurations, including models trained from scratch from small amounts of raw text and models pre-trained on other languages but fine-tuned on Medieval French data.",
}
```