MoE-LPR: Multilingual Extension of Large Language Models through Mixture-of-Experts with Language Priors Routing
- URL: http://arxiv.org/abs/2408.11396v1
- Date: Wed, 21 Aug 2024 07:43:49 GMT
- Title: MoE-LPR: Multilingual Extension of Large Language Models through Mixture-of-Experts with Language Priors Routing
- Authors: Hao Zhou, Zhijun Wang, Shujian Huang, Xin Huang, Xue Han, Junlan Feng, Chao Deng, Weihua Luo, Jiajun Chen,
- Abstract summary: Large Language Models (LLMs) are often English-centric due to the disproportionate distribution of languages in their pre-training data.
We propose a method called MoE-LPR (Mixture-of-Experts with Language Priors) to enhance the multilingual capability.
- Score: 78.62611800987817
- License: http://creativecommons.org/licenses/by/4.0/
- Abstract: Large Language Models (LLMs) are often English-centric due to the disproportionate distribution of languages in their pre-training data. Enhancing non-English language capabilities through post-pretraining often results in catastrophic forgetting of the ability of original languages. Previous methods either achieve good expansion with severe forgetting or slight forgetting with poor expansion, indicating the challenge of balancing language expansion while preventing forgetting. In this paper, we propose a method called MoE-LPR (Mixture-of-Experts with Language Priors Routing) to alleviate this problem. MoE-LPR employs a two-stage training approach to enhance the multilingual capability. First, the model is post-pretrained into a Mixture-of-Experts (MoE) architecture by upcycling, where all the original parameters are frozen and new experts are added. In this stage, we focus improving the ability on expanded languages, without using any original language data. Then, the model reviews the knowledge of the original languages with replay data amounting to less than 1% of post-pretraining, where we incorporate language priors routing to better recover the abilities of the original languages. Evaluations on multiple benchmarks show that MoE-LPR outperforms other post-pretraining methods. Freezing original parameters preserves original language knowledge while adding new experts preserves the learning ability. Reviewing with LPR enables effective utilization of multilingual knowledge within the parameters. Additionally, the MoE architecture maintains the same inference overhead while increasing total model parameters. Extensive experiments demonstrate MoE-LPR's effectiveness in improving expanded languages and preserving original language proficiency with superior scalability. Code and scripts are freely available at https://github.com/zjwang21/MoE-LPR.git.
Related papers
- LangSAMP: Language-Script Aware Multilingual Pretraining [48.16511046793275]
Recent multilingual pretrained language models (mPLMs) often avoid using language embeddings.
LangSAMP incorporates both language and script embeddings to enhance representation learning.
We apply LangSAMP to the continual pretraining of XLM-R on a highly multilingual corpus covering more than 500 languages.
arXiv Detail & Related papers (2024-09-26T18:29:10Z) - Crosslingual Capabilities and Knowledge Barriers in Multilingual Large Language Models [62.91524967852552]
Large language models (LLMs) are typically multilingual due to pretraining on diverse multilingual corpora.
But can these models relate corresponding concepts across languages, effectively being crosslingual?
This study evaluates six state-of-the-art LLMs on inherently crosslingual tasks.
arXiv Detail & Related papers (2024-06-23T15:15:17Z) - Enhancing Multilingual Capabilities of Large Language Models through
Self-Distillation from Resource-Rich Languages [60.162717568496355]
Large language models (LLMs) have been pre-trained on multilingual corpora.
Their performance still lags behind in most languages compared to a few resource-rich languages.
arXiv Detail & Related papers (2024-02-19T15:07:32Z) - Soft Language Clustering for Multilingual Model Pre-training [57.18058739931463]
We propose XLM-P, which contextually retrieves prompts as flexible guidance for encoding instances conditionally.
Our XLM-P enables (1) lightweight modeling of language-invariant and language-specific knowledge across languages, and (2) easy integration with other multilingual pre-training methods.
arXiv Detail & Related papers (2023-06-13T08:08:08Z) - Continual Learning in Multilingual NMT via Language-Specific Embeddings [92.91823064720232]
It consists in replacing the shared vocabulary with a small language-specific vocabulary and fine-tuning the new embeddings on the new language's parallel data.
Because the parameters of the original model are not modified, its performance on the initial languages does not degrade.
arXiv Detail & Related papers (2021-10-20T10:38:57Z) - Multilingual Transfer Learning for QA Using Translation as Data
Augmentation [13.434957024596898]
We explore strategies that improve cross-lingual transfer by bringing the multilingual embeddings closer in the semantic space.
We propose two novel strategies, language adversarial training and language arbitration framework, which significantly improve the (zero-resource) cross-lingual transfer performance.
Empirically, we show that the proposed models outperform the previous zero-shot baseline on the recently introduced multilingual MLQA and TyDiQA datasets.
arXiv Detail & Related papers (2020-12-10T20:29:34Z) - Reusing a Pretrained Language Model on Languages with Limited Corpora
for Unsupervised NMT [129.99918589405675]
We present an effective approach that reuses an LM that is pretrained only on the high-resource language.
The monolingual LM is fine-tuned on both languages and is then used to initialize a UNMT model.
Our approach, RE-LM, outperforms a competitive cross-lingual pretraining model (XLM) in English-Macedonian (En-Mk) and English-Albanian (En-Sq)
arXiv Detail & Related papers (2020-09-16T11:37:10Z)
This list is automatically generated from the titles and abstracts of the papers in this site.
This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.