paper-with-me

홈 › Papers

BERT-of-Theseus: Compressing BERT by Progressive Module Replacing

2020-02-07 · EMNLP 2020 11 · Canwen Xu, Wangchunshu Zhou, Tao Ge, Furu Wei, Ming Zhou

In this paper, we propose a novel model compression approach to effectively compress BERT by progressive module replacing. Our approach first divides the original BERT into several modules and builds their compact substitutes. Then, we randomly replace the original modules with their substitutes to train the compact modules to mimic the behavior of the original modules. We progressively increase the probability of replacement through the training. In this way, our approach brings a deeper level of interaction between the original and compact models. Compared to the previous knowledge distillation approaches for BERT compression, our approach does not introduce any additional loss function. Our approach outperforms existing knowledge distillation approaches on GLUE benchmark, showing a new perspective of model compression.

📄 PDF Abstract BibTeX arXiv:2002.02925

Code (2)

JetRunner/BERT-of-Theseus 공식 구현 pytorch
ambroggi/pruning-project-for-deep-neural-networks pytorch

Tasks

Knowledge DistillationModel Compression

Methods 이 논문이 사용한 방법론

Linear Layer A Linear Layer is a projection $\mathbf{XW + b}$.
Knowledge Distillation A very simple way to improve the performance of almost any machine learning algorithm is to train many different models on the same data and then to average their predictions.…
Residual Connection 설명 없음
Attention Dropout Attention Dropout is a type of dropout used in attention-based architectures, where elements are randomly dropped out of the…
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.
Weight Decay 설명 없음
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 &…
Dense Connections Dense Connections, or Fully Connected Connections, are a type of layer in a deep neural network that use a linear operation where every input is connected to every output…

Similar Papers 제목 키워드 기반

KroneckerBERT: Significant Compression of Pre-trained Language Models Through Kronecker Decomposition and Knowledge Distillation

2022-07-01 · NAACL 2022 7 · Marzieh Tahaei, Ella Charlaix, Vahid Nia, Ali Ghodsi 외

The development of over-parameterized pre-trained language models has made a significant contribution toward the success of natural language processing. While over-parameterization of these models is the key to their gen…

Knowledge DistillationLanguage ModelingLanguage ModellingModel Compression

BERT-JAM: Boosting BERT-Enhanced Neural Machine Translation with Joint Attention

2020-11-09 · Zhebin Zhang, Sai Wu, Dawei Jiang, Gang Chen

BERT-enhanced neural machine translation (NMT) aims at leveraging BERT-encoded representations for translation tasks. A recently proposed approach uses attention mechanisms to fuse Transformer's encoder and decoder layer…

DecoderMachine TranslationNMTTranslation

RefBERT: Compressing BERT by Referencing to Pre-computed Representations

2021-06-11 · Xinyi Wang, Haiqin Yang, Liang Zhao, Yang Mo 외

Recently developed large pre-trained language models, e.g., BERT, have achieved remarkable performance in many downstream natural language processing applications. These pre-trained language models often contain hundreds…

Knowledge Distillation

Compressing Large-Scale Transformer-Based Models: A Case Study on BERT

2020-02-27 · Prakhar Ganesh, Yao Chen, Xin Lou, Mohammad Ali Khan 외

Pre-trained Transformer-based models have achieved state-of-the-art performance for various Natural Language Processing (NLP) tasks. However, these models often have billions of parameters, and, thus, are too resource-hu…

Model Compression

Network of Theseus (like the ship)

2025-12-03 · Vighnesh Subramaniam, Colin Conwell, Boris Katz, Andrei Barbu 외 arxiv

A standard assumption in deep learning is that the inductive bias introduced by a neural network architecture must persist from training through inference. The architecture you train with is the architecture you deploy. …