Omartificial-Intelligence-Space

/

SHAMI-MT-2MSA

like 1

This PR significantly enhances the model card for Omartificial-Intelligence-Space/Shami-MT-2MSA by:

• Adding pipeline_tag: translation to the metadata, which is crucial for model discoverability on the Hugging Face Hub, accurately reflecting its machine translation capabilities.
• Correcting a typo in the main model card title ("ton" to "to").
• Including a direct link to the accompanying paper, SHAMI-MT: A Syrian Arabic Dialect to Modern Standard Arabic Bidirectional Machine Translation System, at the top of the card.
• Adding the paper's abstract to provide comprehensive context about the model's design, training, and evaluation.
• Clarifying the model description to explicitly state that SHAMI-MT-2MSA is the Syrian dialect to MSA component of the broader SHAMI-MT bidirectional system.
• Adding a Python sample usage snippet, demonstrating how to load and use the model with the transformers library, which improves immediate usability.
• Refining the citation section to include a proper BibTeX entry for the main paper and removing redundant details.
• Removing the "Model Details" section as its content is either covered by the metadata or the main description.

These improvements make the model card more informative, user-friendly, and aligned with best practices on the Hugging Face Hub.