mmBERT: A Modern Multilingual Encoder with Annealed Language Learning
Abstract
mmBERT, an encoder-only language model pretrained on multilingual text, achieves high performance on classification and retrieval tasks using an inverse mask ratio schedule and inverse temperature sampling ratio, particularly benefiting from the inclusion of low-resource languages.
Encoder-only languages models are frequently used for a variety of standard machine learning tasks, including classification and retrieval. However, there has been a lack of recent research for encoder models, especially with respect to multilingual models. We introduce mmBERT, an encoder-only language model pretrained on 3T tokens of multilingual text in over 1800 languages. To build mmBERT we introduce several novel elements, including an inverse mask ratio schedule and an inverse temperature sampling ratio. We add over 1700 low-resource languages to the data mix only during the decay phase, showing that it boosts performance dramatically and maximizes the gains from the relatively small amount of training data. Despite only including these low-resource languages in the short decay phase we achieve similar classification performance to models like OpenAI's o3 and Google's Gemini 2.5 Pro. Overall, we show that mmBERT significantly outperforms the previous generation of models on classification and retrieval tasks -- on both high and low-resource languages.
Community
Models and pre-training data available here: https://huggingface.co/collections/jhu-clsp/mmbert-a-modern-multilingual-encoder-68b725831d7c6e3acc435ed4
This is an automated message from the Librarian Bot. I found the following papers similar to this paper.
The following papers were recommended by the Semantic Scholar API
- Utilizing Multilingual Encoders to Improve Large Language Models for Low-Resource Languages (2025)
- MERLIN: Multi-Stage Curriculum Alignment for Multilingual Encoder and LLM Fusion (2025)
- Learning Dynamics of Meta-Learning in Small Model Pretraining (2025)
- Cetvel: A Unified Benchmark for Evaluating Language Understanding, Generation and Cultural Capacity of LLMs for Turkish (2025)
- On The Role of Pretrained Language Models in General-Purpose Text Embeddings: A Survey (2025)
- Llama-GENBA-10B: A Trilingual Large Language Model for German, English and Bavarian (2025)
- CONCAP: Seeing Beyond English with Concepts Retrieval-Augmented Captioning (2025)
Please give a thumbs up to this comment if you found it helpful!
If you want recommendations for any Paper on Hugging Face checkout this Space
You can directly ask Librarian Bot for paper recommendations by tagging it in a comment:
@librarian-bot
recommend
Models citing this paper 3
Datasets citing this paper 13
Browse 13 datasets citing this paperSpaces citing this paper 0
No Space linking this paper