Blockwise Self-Attention for Long Document Understanding
This addresses the computational inefficiency of processing long documents for NLP tasks, offering a more scalable solution, though it is incremental as it builds on existing BERT architectures.
The paper tackles the problem of modeling long-distance dependencies in documents by introducing BlockBERT, a BERT model with sparse block attention structures, which reduces memory usage by 18.7-36.1% and training/inference time by up to 27.8% while maintaining or improving accuracy compared to RoBERTa.
We present BlockBERT, a lightweight and efficient BERT model for better modeling long-distance dependencies. Our model extends BERT by introducing sparse block structures into the attention matrix to reduce both memory consumption and training/inference time, which also enables attention heads to capture either short- or long-range contextual information. We conduct experiments on language model pre-training and several benchmark question answering datasets with various paragraph lengths. BlockBERT uses 18.7-36.1% less memory and 12.0-25.1% less time to learn the model. During testing, BlockBERT saves 27.8% inference time, while having comparable and sometimes better prediction accuracy, compared to an advanced BERT-based model, RoBERTa.