paper-with-me

홈 › Papers

Revisiting Token Dropping Strategy in Efficient BERT Pretraining

2023-05-24 · Qihuang Zhong, Liang Ding, Juhua Liu, Xuebo Liu, Min Zhang, Bo Du, DaCheng Tao

Token dropping is a recently-proposed strategy to speed up the pretraining of masked language models, such as BERT, by skipping the computation of a subset of the input tokens at several middle layers. It can effectively reduce the training time without degrading much performance on downstream tasks. However, we empirically find that token dropping is prone to a semantic loss problem and falls short in handling semantic-intense tasks. Motivated by this, we propose a simple yet effective semantic-consistent learning method (ScTD) to improve the token dropping. ScTD aims to encourage the model to learn how to preserve the semantic information in the representation space. Extensive experiments on 12 tasks show that, with the help of our ScTD, token dropping can achieve consistent and significant performance gains across all task types and model sizes. More encouragingly, ScTD saves up to 57% of pretraining time and brings up to +1.56% average improvement over the vanilla token dropping.

📄 PDF Abstract BibTeX arXiv:2305.15273

Code (1)

whu-zqh/sctd 공식 구현 pytorch

Methods 이 논문이 사용한 방법론

Linear Warmup With Linear Decay Linear Warmup With Linear Decay is a learning rate schedule in which we increase the learning rate linearly for $n$ updates and then linearly decay afterwards.
Refunds@Expedia|||How do I get a full refund from Expedia? “How do I get a full refund from Expedia? How do I get a full refund from Expedia? – Call ☎️ +1-(888) 829 (0881) or +1-805-330-4056 or +1-805-330-4056 for Quick Help &…
WordPiece 설명 없음
Softmax The Softmax output function transforms a previous layer's output into a vector of probabilities. It is commonly used for multiclass classification. Given an input vector $x$…
Layer Normalization Unlike batch normalization, Layer Normalization directly estimates the normalization statistics from the summed inputs…
Dropout Dropout is a regularization technique for neural networks that drops a unit (along with connections) at training time with a specified probability $p$ (a common value is…
Attention 설명 없음
Linear Layer A Linear Layer is a projection $\mathbf{XW + b}$.

Similar Papers 제목 키워드 기반

Token Dropping for Efficient BERT Pretraining

2022-03-24 · ACL 2022 5 · Le Hou, Richard Yuanzhe Pang, Tianyi Zhou, Yuexin Wu 외

Transformer-based models generally allocate the same amount of computation for each token in a given sequence. We develop a simple but effective "token dropping" method to accelerate the pretraining of transformer models…

Language ModelingLanguage ModellingMasked Language Modeling

Random-LTD: Random and Layerwise Token Dropping Brings Efficient Training for Large-scale Transformers

2022-11-17 · Zhewei Yao, Xiaoxia Wu, Conglong Li, Connor Holmes 외

Large-scale transformer models have become the de-facto architectures for various machine learning applications, e.g., CV and NLP. However, those large models also introduce prohibitive training costs. To mitigate this i…

moBERTo: A Modern Encoder for Portuguese via Continued Pretraining of ModernBERT

2026-06-21 · Thiago Laitz, Thales Sales Almeida, João Guilherme Alves Santos, Giovana Kerche Bonás arxiv

Encoder-only transformer models remain essential for production NLP pipelines. We introduce moBERTo, a Portuguese adaptation of ModernBERT obtained through continued pretraining of the ModernBERT-base checkpoint on 60 bi…

Natural Language UnderstandingDocument ClassificationInformation Retrieval

NarrowBERT: Accelerating Masked Language Model Pretraining and Inference

2023-01-11 · Haoxin Li, Phillip Keung, Daniel Cheng, Jungo Kasai 외

Large-scale language model pretraining is a very successful form of self-supervised learning in natural language processing, but it is increasingly expensive to perform as the models and pretraining corpora have become l…

Language ModelingLanguage ModellingmodelNER+2

Efficient Stagewise Pretraining via Progressive Subnetworks

2024-02-08 · Abhishek Panigrahi, Nikunj Saunshi, Kaifeng Lyu, Sobhan Miryoosefi 외

Recent developments in large language models have sparked interest in efficient pretraining methods. Stagewise training approaches to improve efficiency, like gradual stacking and layer dropping (Reddi et al, 2023; Zhang…

Inductive Bias