Aurora-M: Open Source Continual Pre-training for Multilingual Language and Code
Fuente:
arXiv
Gespeichert in:
| Hauptverfasser: | , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , |
|---|---|
| Format: | Preprint |
| Veröffentlicht: |
2024
|
| Schlagworte: | |
| Online-Zugang: | |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| _version_ | 1866915079386038272 |
|---|---|
| author | Nakamura, Taishi Mishra, Mayank Tedeschi, Simone Chai, Yekun Stillerman, Jason T Friedrich, Felix Yadav, Prateek Laud, Tanmay Chien, Vu Minh Zhuo, Terry Yue Misra, Diganta Bogin, Ben Vu, Xuan-Son Karpinska, Marzena Dantuluri, Arnav Varma Kusa, Wojciech Furlanello, Tommaso Yokota, Rio Muennighoff, Niklas Pai, Suhas Adewumi, Tosin Laippala, Veronika Yao, Xiaozhe Junior, Adalberto Ariyak, Alpay Drozd, Aleksandr Clive, Jordan Gupta, Kshitij Chen, Liangyu Sun, Qi Tsui, Ken Persaud, Noah Fahmy, Nour Chen, Tianlong Bansal, Mohit Monti, Nicolo Dang, Tai Luo, Ziyang Bui, Tien-Tung Navigli, Roberto Mehta, Virendra Blumberg, Matthew May, Victor Nguyen, Huu Pyysalo, Sampo |
| author_facet | Nakamura, Taishi Mishra, Mayank Tedeschi, Simone Chai, Yekun Stillerman, Jason T Friedrich, Felix Yadav, Prateek Laud, Tanmay Chien, Vu Minh Zhuo, Terry Yue Misra, Diganta Bogin, Ben Vu, Xuan-Son Karpinska, Marzena Dantuluri, Arnav Varma Kusa, Wojciech Furlanello, Tommaso Yokota, Rio Muennighoff, Niklas Pai, Suhas Adewumi, Tosin Laippala, Veronika Yao, Xiaozhe Junior, Adalberto Ariyak, Alpay Drozd, Aleksandr Clive, Jordan Gupta, Kshitij Chen, Liangyu Sun, Qi Tsui, Ken Persaud, Noah Fahmy, Nour Chen, Tianlong Bansal, Mohit Monti, Nicolo Dang, Tai Luo, Ziyang Bui, Tien-Tung Navigli, Roberto Mehta, Virendra Blumberg, Matthew May, Victor Nguyen, Huu Pyysalo, Sampo |
| contents | Pretrained language models are an integral part of AI applications, but their high computational cost for training limits accessibility. Initiatives such as Bloom and StarCoder aim to democratize access to pretrained models for collaborative community development. Despite these efforts, such models encounter challenges such as limited multilingual capabilities, risks of catastrophic forgetting during continual pretraining, and the high costs of training models from scratch, alongside the need to align with AI safety standards and regulatory frameworks.
This paper presents Aurora-M, a 15B parameter multilingual open-source model trained on English, Finnish, Hindi, Japanese, Vietnamese, and code. Continually pretrained from StarCoderPlus on 435B additional tokens, Aurora-M surpasses 2T tokens in total training token count. It is the first open-source multilingual model fine-tuned on human-reviewed safety instructions, thus aligning its development not only with conventional red-teaming considerations, but also with the specific concerns articulated in the Biden-Harris Executive Order on the Safe, Secure, and Trustworthy Development and Use of Artificial Intelligence.
We evaluate Aurora-M across a wide range of tasks and languages, showcasing its robustness against catastrophic forgetting and its superior performance in multilingual settings, particularly in safety evaluations. We open-source Aurora-M and its variants to encourage responsible open-source development of large language models at https://huggingface.co/aurora-m. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2404_00399 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | Aurora-M: Open Source Continual Pre-training for Multilingual Language and Code Nakamura, Taishi Mishra, Mayank Tedeschi, Simone Chai, Yekun Stillerman, Jason T Friedrich, Felix Yadav, Prateek Laud, Tanmay Chien, Vu Minh Zhuo, Terry Yue Misra, Diganta Bogin, Ben Vu, Xuan-Son Karpinska, Marzena Dantuluri, Arnav Varma Kusa, Wojciech Furlanello, Tommaso Yokota, Rio Muennighoff, Niklas Pai, Suhas Adewumi, Tosin Laippala, Veronika Yao, Xiaozhe Junior, Adalberto Ariyak, Alpay Drozd, Aleksandr Clive, Jordan Gupta, Kshitij Chen, Liangyu Sun, Qi Tsui, Ken Persaud, Noah Fahmy, Nour Chen, Tianlong Bansal, Mohit Monti, Nicolo Dang, Tai Luo, Ziyang Bui, Tien-Tung Navigli, Roberto Mehta, Virendra Blumberg, Matthew May, Victor Nguyen, Huu Pyysalo, Sampo Computation and Language Artificial Intelligence Machine Learning Pretrained language models are an integral part of AI applications, but their high computational cost for training limits accessibility. Initiatives such as Bloom and StarCoder aim to democratize access to pretrained models for collaborative community development. Despite these efforts, such models encounter challenges such as limited multilingual capabilities, risks of catastrophic forgetting during continual pretraining, and the high costs of training models from scratch, alongside the need to align with AI safety standards and regulatory frameworks. This paper presents Aurora-M, a 15B parameter multilingual open-source model trained on English, Finnish, Hindi, Japanese, Vietnamese, and code. Continually pretrained from StarCoderPlus on 435B additional tokens, Aurora-M surpasses 2T tokens in total training token count. It is the first open-source multilingual model fine-tuned on human-reviewed safety instructions, thus aligning its development not only with conventional red-teaming considerations, but also with the specific concerns articulated in the Biden-Harris Executive Order on the Safe, Secure, and Trustworthy Development and Use of Artificial Intelligence. We evaluate Aurora-M across a wide range of tasks and languages, showcasing its robustness against catastrophic forgetting and its superior performance in multilingual settings, particularly in safety evaluations. We open-source Aurora-M and its variants to encourage responsible open-source development of large language models at https://huggingface.co/aurora-m. |
| title | Aurora-M: Open Source Continual Pre-training for Multilingual Language and Code |
| topic | Computation and Language Artificial Intelligence Machine Learning |
| url | https://arxiv.org/abs/2404.00399 |