Downloads · 30 days
11
3% of all-time downloads
mdeberta-v3-base-finetuned-ner
CLARA-MeD/mdeberta-v3-base-finetuned-ner
mdeberta-v3-base-finetuned-ner is a machine learning model from CLARA-MeD. Use it for the machine learning task on the model card, and read the license before you ship it in a product. The card lists the license as cc-by-nc-4.0.
This is a model for complex word identification (CWI) of Spanish medical texts, based on the multilingual DeBERTa vs 3 (mDeBERTa).
All-time downloads
363
Public
Parameters
278M
30.1 GB on disk
Likes
0
Public
Repo makeup
Click a slice to open those files.
.safetensors1.1 GB · 98%
At a glance
- License
- cc-by-nc-4.0
- Model type
- deberta-v2
- Access
- Public
- Created
- Aug 23, 2024
- Updated
- Mar 3, 2025
- SHA
- 2b040622
Model card
From the Hugging Face model README
This is a model for complex word identification (CWI) of Spanish medical texts, based on the multilingual DeBERTa vs 3 (mDeBERTa).
The model was fine-tuned on a corpus of 225 texts for patients (162575 tokens) to identify complex words (CW).
Results (test set)
| Class | Precision | Recall | F1 | Accuracy |
|---|---|---|---|---|
| CW | 79.05 (±1.39) | 79.01 (±0.70) | 79.02 (±0.65) | 94.86 (±0.22) |
*Results are the average of 3 experimental rounds.
If you use this model or want to have more details about the experiments and the training details, take a look at our article:
@article{2025CWI,
title={Complex Word Identification for Lexical Simplification in Spanish Texts for Patients},
author={Ortega-Riba, Federico and Campillos-Llanos, Leonardo and Samy, Doaa},
journal={Procesamiento del lenguaje natural},
volume={74},
year={2025}
}
- Type
- deberta-v2
- License
- cc-by-nc-4.0
- Languages
- es
- Created
- Aug 23, 2024
- Updated
- Mar 3, 2025