himel7/da-multilingual-bias-detector

Model Card for Model ID

This is a model for bias detection in news texts in multiple languages.

This model is distilled from a pre-trained domain adapted and fine tuned bias detection model, hence it transfers the domain-adapted learning approach along with the fine-tuning to a multilingual xlm-roberta base model.

On BABE Test Set in English: Accuracy- 0.826 Macro F1- 0.8324

On BABE Test Set in Hindi: Accuracy- 0.798 Macro F1- 0.8206

On BABE Test Set in Bengali: Accuracy- 0.757 Macro F1- 0.797

You can add further metrics in other languages here...

Model Details

Model Description

This is the model card of a 🤗 transformers model that has been pushed on the Hub. This model card has been automatically generated.

• Developed by: Himel Ghosh

• Language(s) (NLP): English, Italian, German, etc.

• Finetuned from model xlm-roberta: Distilled from mediabiasgroup/da-roberta-babe-ft