Kai-Fu Lee (@kaifulee) started following @LiJunnan0409 on Oct 13, 2024

RT @DongxuLi_: Great thread to learn more about Aria pretraining pipelines and data recipes at @rhymes_ai_

🧵 (5/6) Unlike typical post-training of language models, our post-training involves a large amount of 20B tokens due to the diversity of multimodal data (e.g., documents, charts, videos). Some domains remain uncovered, where we resort to Aria's generalization capabilities.

Great thread to learn more about Aria pretraining pipelines and data recipes at @rhymes_ai_

The Aria model from @rhymes_ai_ is top trending on @huggingface . Also humbled to receive very positive community feedback, and appreciate third-party evaluations. Report: https://t.co/HdQXTDqeqG Model on HF: https://t.co/8N83V9dw5y https://t.co/VvE0J5kjGt