From the source

Falcon-Emirati: When an LLM Learns the Dialect, the Culture, and the Nuance

Falcon-Emirati-7B adapts a general Arabic model to the Emirati dialect and culture.

New model

Image published with the post by Hugging Face — Image from huggingface.co

TII released Falcon-Emirati-7B, a dialect-specialized model built on Falcon-H1-Arabic that understands and generates Emirati Arabic with cultural context.

The model uses the Falcon-H1 hybrid architecture combining Mamba and Transformer attention, and was trained on authentic Emirati web data, MSA cultural content, and synthetic data guided by glossaries.

The 7B parameter variant was chosen as the best balance of quality and cost for dialect adaptation.

From the source
Falcon-Emirati-7B is built on Falcon-H1-Arabic, our Arabic model family that already set new benchmarks for the language earlier this year. Falcon-H1-Arabic uses the Falcon-H1 hybrid architecture: State Space Models (Mamba) and Transformer attention running in parallel inside every block, with their outputs fused before each block's projection.
huggingface.co

Other stories

Other stories

Falcon-Emirati: When an LLM Learns the Dialect, the Culture, and the Nuance

From the source

Other stories