NarrowBERT: Accelerating Masked Language Model Pretraining and Inference (2023.acl-short)
Copied to clipboard
| Challenge: | Large-scale language model pretraining is expensive as the models and pretraining corpora have become larger over time. |
| Approach: | They propose a modified transformer encoder that increases throughput for masked language model pretraining by more than 2x. |
| Outcome: | The proposed model increases throughput on IMDB and Amazon reviews classification and CoNLL NER tasks by 3.5x with minimal performance degradation. |
Similar Papers
HybridBERT - Making BERT Pretraining More Efficient Through Hybrid Mixture of Attention Mechanisms (2024.naacl-srw)
Copied to clipboard
| Challenge: | Pretrained transformer-based language models have produced state-of-the-art performance in most natural language understanding tasks. |
| Approach: | They propose two hybrid architectures that combine self-attention and additive attention mechanisms with sub-layer normalization to achieve double the pretraining accuracy of a vanilla-BERT baseline. |
| Outcome: | The proposed architectures outperform BERT-base on two downstream tasks while accelerating inference. |
The Diminishing Returns of Masked Language Models to Science (2023.findings-acl)
Copied to clipboard
| Challenge: | Existing studies have shown that masked language models can improve downstream tasks by pretraining larger models for longer on more data. |
| Approach: | They empirically evaluate the extent to which these results extend to tasks in science by using 14 domain-specific transformer-based masked language models. |
| Outcome: | The proposed model can improve on 12 scientific tasks, but not all. |
DecBERT: Enhancing the Language Understanding of BERT with Causal Attention Masks (2022.findings-naacl)
Copied to clipboard
| Challenge: | Experimental results show that Transformer Encoder model can't automatically capture word order, so explicit position embeddings are required to be fed into the target model. |
| Approach: | They propose a Transformer-based language model DecBERT that uses a causal attention mask to capture word order. |
| Outcome: | The proposed model improves on the GLUE language understanding benchmark and accelerates the pre-training process. |
Sentence Bottleneck Autoencoders from Transformer Language Models (2021.emnlp-main)
Copied to clipboard
| Challenge: | Existing methods for pretraining a language model on text have been used for building models in NLP, but they do not work for sentence representations derived from pretrainer models based on tokens or basic pooling operations. |
| Approach: | They propose to build a sentence-level autoencoder from a pretrained transformer language model. |
| Outcome: | The proposed model achieves better quality than previous methods on text similarity and style transfer tasks while using fewer parameters than large pretrained models. |
Training compute-optimal transformer encoder models (2025.emnlp-main)
Copied to clipboard
| Challenge: | OptiBERT is a family of compute-optimal BERT-style models that matches or surpasses leading baselines while training with dramatically less FLOPS. |
| Approach: | They propose to train OptiBERT models with a Masked Language Model objective . they train a family of compute-optimal BERT-style models that matches or surpasses leading baselines . |
| Outcome: | The proposed model matches or surpasses leading baselines on GLUE and MTEB while training with dramatically less FLOPS. |
DiffusionBERT: Improving Generative Masked Language Models with Diffusion Models (2023.acl-long)
Copied to clipboard
| Challenge: | Existing generative masked language models have a shared training objective, i.e., denoising. |
| Approach: | They propose a noise schedule for the forward diffusion process that controls the degree of noise added at each step based on the information of each token. |
| Outcome: | The proposed model improves on existing models in terms of perplexity and BLEU score. |
TinyBERT: Distilling BERT for Natural Language Understanding (2020.findings-emnlp)
Copied to clipboard
| Challenge: | Pre-trained language models are computationally expensive and difficult to efficiently execute on resource-restricted devices. |
| Approach: | They propose a Transformer distillation method that performs Transformer distillations at pre-training and task-specific learning stages. |
| Outcome: | The proposed method accelerates inference and reduces model size while maintaining accuracy. |
Revisiting Pre-Trained Models for Chinese Natural Language Processing (2020.findings-emnlp)
Copied to clipboard
| Challenge: | Existing pre-trained language models have shown tremendous improvements across various NLP tasks. |
| Approach: | They propose to revisit Chinese pre-trained language models to examine their effectiveness in a non-English language and release the Chinese pretrained model series to the community. |
| Outcome: | The proposed model improves on RoBERTa in several ways, especially the masking strategy that adopts MLM as correction (Mac). |
LinkBERT: Pretraining Language Models with Document Links (2022.acl-long)
Copied to clipboard
| Challenge: | Existing language model pretraining methods do not capture dependencies or knowledge that span across documents. |
| Approach: | They propose a language model pretraining method that leverages links between documents . they use masked language modeling and document relation prediction to model LMs . |
| Outcome: | The proposed method outperforms existing methods on downstream tasks across two domains. |
Masking as an Efficient Alternative to Finetuning for Pretrained Language Models (2020.emnlp-main)
Copied to clipboard
| Challenge: | Extensive evaluations of masking BERT, RoBERTa, and DistilBERT on eleven diverse NLP tasks show that our binary masked language models encode information necessary for solving downstream tasks. |
| Approach: | They propose an efficient method of utilizing pretrained language models where selective binary masks are learned instead of finetuning. |
| Outcome: | Extensive evaluations of masking BERT, RoBERTa, and DistilBERT on eleven diverse NLP tasks show that the proposed method yields comparable performance to finetuning, but has a much smaller memory footprint when multiple tasks need to be solved. |