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. |
Similar Papers
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. |
EarlyBERT: Efficient BERT Training via Early-bird Lottery Tickets (2021.acl-long)
Copied to clipboard
| Challenge: | Large-scale pre-trained language models require enormous computational resources and long training time. |
| Approach: | They propose an algorithm to reduce inference time and train large NLP models by slimming the self-attention and fully-connected sub-layers inside a transformer. |
| Outcome: | The proposed algorithm achieves comparable performance to standard BERT with 35 45% less training time. |
bert2BERT: Towards Reusable Pretrained Language Models (2022.acl-long)
Copied to clipboard
Cheng Chen, Yichun Yin, Lifeng Shang, Xin Jiang, Yujia Qin, Fengyu Wang, Zhi Wang, Xiao Chen, Zhiyuan Liu, Qun Liu
| Challenge: | Pre-training large language models can be expensive and wasteful. |
| Approach: | They propose a method which can transfer the knowledge of an existing smaller pre-trained model to a large model through parameter initialization and a two-stage learning method to further accelerate the pre-training. |
| Outcome: | The proposed method can transfer the knowledge of an existing smaller pre-trained model to a large model through parameter initialization and significantly improve the pre-training efficiency of the large model. |
DABERT: Dual Attention Enhanced BERT for Semantic Matching (2022.coling-1)
Copied to clipboard
| Challenge: | Existing models for semantic sentence matching lack the ability to capture subtle differences. |
| Approach: | They propose to use a Transformer-based pre-trained language model to capture fine-grained differences in sentence pairs by introducing a dual attention module and a fusion module to learn the aggregation of difference and affinity features. |
| Outcome: | The proposed method is able to capture fine-grained differences in sentence pairs. |
Attention Fusion: a light yet efficient late fusion mechanism for task adaptation in NLU (2022.findings-naacl)
Copied to clipboard
| Challenge: | a recent study has shown that fine-tuning pre-trained models is parameter-inefficient and expensive. |
| Approach: | They propose a task-attuned token module which integrates pre-trained network representations into a pre-trainer. |
| Outcome: | The proposed model trains only 0.0009% of the parameters and is efficient during computation and scalable during deployment. |
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. |
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. |
MoEBERT: from BERT to Mixture-of-Experts via Importance-Guided Adaptation (2022.naacl-main)
Copied to clipboard
| Challenge: | Existing methods for training pre-trained language models have limited practicality due to latency requirements. |
| Approach: | They propose a method that uses a Mixture-of-Experts structure to increase model capacity and inference speed. |
| Outcome: | The proposed method outperforms existing distillation methods on natural language understanding and question answering tasks. |
Taming Pre-trained Language Models with N-gram Representations for Low-Resource Domain Adaptation (2021.acl-long)
Copied to clipboard
| Challenge: | Existing methods to train pre-trained models require domain-specific data and computational resources. |
| Approach: | They propose a domain-aware N-gram Adaptor to incorporate unseen and domain-specific words into a generic pretrained model. |
| Outcome: | The proposed model can improve on eight low-resource tasks using limited data with lower computational costs. |
DILBERT: Customized Pre-Training for Domain Adaptation with Category Shift, with an Application to Aspect Extraction (2021.emnlp-main)
Copied to clipboard
| Challenge: | Existing methods for pre-training can be sub-optimal in some cases . for example, aspect extraction tasks require domain and category invariant representations . |
| Approach: | They propose a domain-invariant learning scheme for BERT to fine-tune pre-trained language models on a source domain and then apply it to a different target domain. |
| Outcome: | The proposed scheme improves performance over state-of-the-art models while using fraction of the unlabeled data. |