Tweeties in a Tweety World

community

AI & ML interests

Multilingual and Low-Resource NLP

Organization Card

The Tweeties is a series of foundation models incorporating native tokenizers for each language, for a better understanding and generation of text in these languages. These models are adapted from existing models using trans-tokenization, and further pre-trained on existing corpora.

Collections 1

models 6

datasets 0

None public yet

Tweeties in a Tweety World

AI & ML interests

Collections 1

Trans-Tokenization and Cross-lingual Vocabulary Transfers: Language Adaptation of LLMs for Low-Resource NLP

Trans-Tokenization and Cross-lingual Vocabulary Transfers: Language Adaptation of LLMs for Low-Resource NLP

models 6

Tweeties/tweety-7b-dutch-v24a

Tweeties/tweety-tatar-hydra-mt-7b-v24a

Tweeties/tweety-tatar-hydra-base-7b-v24a

Tweeties/tweety-7b-tatar-v24a

Tweeties/tweety-7b-armenian-v24a

Tweeties/tweety-7b-italian-v24a

datasets 0

AI & ML interests

Team members 7

Collections 1

models 6 Sort: Recently updated

datasets 0

models 6