Cover Image for Alibaba's Multilingual LLM, Babel, Designed to Serve Over 90% of Global Speakers
Cover Image for Alibaba's Multilingual LLM, Babel, Designed to Serve Over 90% of Global Speakers
Avatar for BuzzRobot
Presented by
BuzzRobot
AI research discussions
Hosted By
69 Going

Alibaba's Multilingual LLM, Babel, Designed to Serve Over 90% of Global Speakers

Zoom
Registration
Welcome! To join the event, please register below.
About Event

Our guest Yiran Zhao from the National University of Singapore will talk about the LLM he worked on with the Alibaba's AI research team.

They've built Babel, an open multilingual LLM that covers the top 25 languages by number of speakers, supports over 90% of the global population, and includes many languages that are not represented in other open multilingual LLMs.

They introduce two variants: Babel-9B, designed for efficient inference and fine-tuning, and Babel-83B.

Evaluations on multilingual tasks demonstrate its better performance compared to open LLMs of comparable size. 

The paper

Join BuzzRobot Slack to connect with the community

Avatar for BuzzRobot
Presented by
BuzzRobot
AI research discussions
Hosted By
69 Going