paper-with-me

홈 › Papers

Allocating Large Vocabulary Capacity for Cross-lingual Language Model Pre-training

2021-09-15 · EMNLP 2021 11 · Bo Zheng, Li Dong, Shaohan Huang, Saksham Singhal, Wanxiang Che, Ting Liu, Xia Song, Furu Wei

Compared to monolingual models, cross-lingual models usually require a more expressive vocabulary to represent all languages adequately. We find that many languages are under-represented in recent cross-lingual language models due to the limited vocabulary capacity. To this end, we propose an algorithm VoCap to determine the desired vocabulary capacity of each language. However, increasing the vocabulary size significantly slows down the pre-training speed. In order to address the issues, we propose k-NN-based target sampling to accelerate the expensive softmax. Our experiments show that the multilingual vocabulary learned with VoCap benefits cross-lingual language model pre-training. Moreover, k-NN-based target sampling mitigates the side-effects of increasing the vocabulary size while achieving comparable performance and faster pre-training speed. The code and the pretrained multilingual vocabularies are available at https://github.com/bozheng-hit/VoCapXLM.

📄 PDF Abstract BibTeX arXiv:2109.07306

Code (2)

bozheng-hit/vocapxlm 공식 구현 pytorch
hit-scir/chinese-mixtral-8x7b pytorch

Tasks

Language ModelingLanguage Modelling

Similar Papers 제목 키워드 기반

XLM-V: Overcoming the Vocabulary Bottleneck in Multilingual Masked Language Models

2023-01-25 · Davis Liang, Hila Gonen, Yuning Mao, Rui Hou 외

Large multilingual language models typically rely on a single vocabulary shared across 100+ languages. As these models have increased in parameter count and depth, vocabulary size has remained largely unchanged. This \te…

Language ModelingLanguage Modellingnamed-entity-recognitionNamed Entity Recognition+4

Towards Continual Learning for Multilingual Machine Translation via Vocabulary Substitution

2021-03-11 · NAACL 2021 4 · Xavier Garcia, Noah Constant, Ankur P. Parikh, Orhan Firat

We propose a straightforward vocabulary adaptation scheme to extend the language capacity of multilingual machine translation models, paving the way towards efficient continual learning for multilingual machine translati…

Continual LearningMachine TranslationTranslation

Focus on the Target's Vocabulary: Masked Label Smoothing for Machine Translation

2022-03-06 · Liang Chen, Runxin Xu, Baobao Chang

Label smoothing and vocabulary sharing are two widely used techniques in neural machine translation models. However, we argue that simply applying both techniques can be conflicting and even leads to sub-optimal performa…

Machine TranslationTranslation

Focus on the Target’s Vocabulary: Masked Label Smoothing for Machine Translation

2022-05-01 · ACL 2022 5 · Liang Chen, Runxin Xu, Baobao Chang

Label smoothing and vocabulary sharing are two widely used techniques in neural machine translation models. However, we argue that simply applying both techniques can be conflicting and even leads to sub-optimal performa…

Machine TranslationTranslation

Overlap-based Vocabulary Generation Improves Cross-lingual Transfer Among Related Languages

2022-03-03 · ACL 2022 5 · Vaidehi Patil, Partha Talukdar, Sunita Sarawagi

Pre-trained multilingual language models such as mBERT and XLM-R have demonstrated great potential for zero-shot cross-lingual transfer to low web-resource languages (LRL). However, due to limited model capacity, the lar…

Cross-Lingual TransferXLM-RZero-Shot Cross-Lingual Transfer