Papers › ConvBERT: Improving BERT with Span-based Dynamic Convolution

ConvBERT: Improving BERT with Span-based Dynamic Convolution

6 Aug 2020NeurIPS 2020 12arXiv:2008.02496archive 2025-07-28

Zi-Hang Jiang, Weihao Yu, Daquan Zhou, Yunpeng Chen, Jiashi Feng, Shuicheng Yan

Pre-trained language models like BERT and its variants have recently achieved impressive performance in various natural language understanding tasks. However, BERT heavily relies on the global self-attention block and thus suffers large memory footprint and computation cost. Although all its attention heads query on the whole input sequence for generating the attention map from a global perspective, we observe some heads only need to learn local dependencies, which means the existence of computation redundancy. We therefore propose a novel span-based dynamic convolution to replace these self-attention heads to directly model local dependencies. The novel convolution heads, together with the rest self-attention heads, form a new mixed attention block that is more efficient at both global and local context learning. We equip BERT with this mixed attention design and build a ConvBERT model. Experiments have shown that ConvBERT significantly outperforms BERT and its variants in various downstream tasks, with lower training cost and fewer model parameters. Remarkably, ConvBERTbase model achieves 86.4 GLUE score, 0.7 higher than ELECTRAbase, while using less than 1/4 training cost. Code and pre-trained models will be released.

PaperPDFConference PDFCode

In Syntology Open this paper in Syntology's Atlas, the map of the papers in Syntology's graph and their citations.

Code

yitu-opensource/ConvBert officialmentioned in papermentioned on GitHubtfNOASSERTION report
JunnYu/paddle_convbert mentioned on GitHubpaddle report
huggingface/transformers mentioned on GitHubpytorch report
stefan-it/europeana-bert mentioned on GitHubtfMIT report
trthackthonFighters/ConvBert mentioned on GitHubtfNOASSERTION report
yyht/Conv_Bert mentioned on GitHubtfNOASSERTION report

Repository list and official/mentioned flags are the archive's, frozen 2025-07-28. Reachability, where shown, is from one Syntology probe window (2026-09-16 to 2026-09-18); repositories not probed show nothing. GitHub stars are not tracked.

Code Syntology ran Syntology

Not run by Syntology. Nothing on this page verifies that the listed code works.

Tasks

Natural Language Understanding

Results from the paper archive 2025-07-28

No leaderboard rows for this paper in the archive.

Methods

Introduced by this paper: ConvBERT, Mixed Attention Block, Span-Based Dynamic Convolution

AdamAttentionAttention DropoutBERTConvBERTConvolutionDense ConnectionsDropoutDynamicConvLayer NormalizationLinear LayerLinear Warmup With Linear DecayMixed Attention BlockMulti-Head AttentionResidual ConnectionSoftmaxSpan-Based Dynamic ConvolutionWeight DecayWordPiece

Report a problem or propose a change · a person checks every report against the paper or source before anything changes; decisions are listed on /corrections