# An Automatic and Efficient BERT Pruning for Edge AI Systems

arXiv: https://arxiv.org/abs/2206.10461 · HTML: https://mlsanity.com/paper/2206.10461/

## Key facts

- Cited by 13 papers in the map (data through 14 Oct 2026).
- Builds on 31 of its 40 references that are in the map.
- First posted on arXiv 21 Jun 2022.

## Builds on (most cited first, 30 of 31)

- [Attention Is All You Need](https://mlsanity.com/paper/1706.03762/) (2017)
- [BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding](https://mlsanity.com/paper/1810.04805/) (2018)
- [An Image is Worth 16x16 Words: Transformers for Image Recognition at Scale](https://mlsanity.com/paper/2010.11929/) (2020)
- [GLUE: A Multi-Task Benchmark and Analysis Platform for Natural Language Understanding](https://mlsanity.com/paper/1804.07461/) (2018)
- [DistilBERT, a distilled version of BERT: smaller, faster, cheaper and lighter](https://mlsanity.com/paper/1910.01108/) (2019)
- [Longformer: The Long-Document Transformer](https://mlsanity.com/paper/2004.05150/) (2020)
- [XLNet: Generalized Autoregressive Pretraining for Language Understanding](https://mlsanity.com/paper/1906.08237/) (2019)
- [The Lottery Ticket Hypothesis: Finding Sparse, Trainable Neural Networks](https://mlsanity.com/paper/1803.03635/) (2018)
- [HuggingFace's Transformers: State-of-the-art Natural Language Processing](https://mlsanity.com/paper/1910.03771/) (2019)
- [Learning both Weights and Connections for Efficient Neural Networks](https://mlsanity.com/paper/1506.02626/) (2015)
- [Reformer: The Efficient Transformer](https://mlsanity.com/paper/2001.04451/) (2020)
- [MiniLM: Deep Self-Attention Distillation for Task-Agnostic Compression of Pre-Trained Transformers](https://mlsanity.com/paper/2002.10957/) (2020)
- [TinyBERT: Distilling BERT for Natural Language Understanding](https://mlsanity.com/paper/1909.10351/) (2019)
- [Unsupervised Data Augmentation for Consistency Training](https://mlsanity.com/paper/1904.12848/) (2019)
- [Effective Approaches to Attention-based Neural Machine Translation](https://mlsanity.com/paper/1508.04025/) (2015)
- [SemEval-2017 Task 1: Semantic Textual Similarity - Multilingual and Cross-lingual Focused Evaluation](https://mlsanity.com/paper/1708.00055/) (2017)
- [What is the State of Neural Network Pruning?](https://mlsanity.com/paper/2003.03033/) (2020)
- [AMC: AutoML for Model Compression and Acceleration on Mobile Devices](https://mlsanity.com/paper/1802.03494/) (2018)
- [The Lottery Ticket Hypothesis for Pre-trained BERT Networks](https://mlsanity.com/paper/2007.12223/) (2020)
- [Compressing BERT: Studying the Effects of Weight Pruning on Transfer Learning](https://mlsanity.com/paper/2002.08307/) (2020)
- [Drawing Early-Bird Tickets: Towards More Efficient Training of Deep Networks](https://mlsanity.com/paper/1909.11957/) (2019)
- [HIBERT: Document Level Pre-training of Hierarchical Bidirectional Transformers for Document Summarization](https://mlsanity.com/paper/1905.06566/) (2019)
- [When BERT Plays the Lottery, All Tickets Are Winning](https://mlsanity.com/paper/2005.00561/) (2020)
- [How Does BERT Answer Questions? A Layer-Wise Analysis of Transformer Representations](https://mlsanity.com/paper/1909.04925/) (2019)
- [AutoCompress: An Automatic DNN Structured Pruning Framework for Ultra-High Compression Rates](https://mlsanity.com/paper/1907.03141/) (2019)
- [Rethinking Network Pruning -- under the Pre-train and Fine-tune Paradigm](https://mlsanity.com/paper/2104.08682/) (2021)
- [Efficient Transformer-based Large Scale Language Representations using Hardware-friendly Block Structured Pruning](https://mlsanity.com/paper/2009.08065/) (2020)
- [Reweighted Proximal Pruning for Large-Scale Language Representation](https://mlsanity.com/paper/1909.12486/) (2019)
- [Accelerating Framework of Transformer by Hardware Design and Model Compression Co-Optimization](https://mlsanity.com/paper/2110.10030/) (2021)
- [C-LSTM: Enabling Efficient LSTM using Structured Compression Techniques on FPGAs](https://mlsanity.com/paper/1803.06305/) (2018)

## Built on by (most cited first, 13 of 13)

- [A Survey of Techniques for Optimizing Transformer Inference](https://mlsanity.com/paper/2307.07982/) (2023)
- [AutoReP: Automatic ReLU Replacement for Fast Private Network Inference](https://mlsanity.com/paper/2308.10134/) (2023)
- [LinGCN: Structural Linearized Graph Convolutional Network for Homomorphically Encrypted Inference](https://mlsanity.com/paper/2309.14331/) (2023)
- [A Length Adaptive Algorithm-Hardware Co-design of Transformer on FPGA Through Sparse Attention and Dynamic Pipelining](https://mlsanity.com/paper/2208.03646/) (2022)
- [Accel-GCN: High-Performance GPU Accelerator Design for Graph Convolution Networks](https://mlsanity.com/paper/2308.11825/) (2023)
- [Towards Sparsification of Graph Neural Networks](https://mlsanity.com/paper/2209.04766/) (2022)
- [Dynamic Sparse Training via Balancing the Exploration-Exploitation Trade-off](https://mlsanity.com/paper/2211.16667/) (2022)
- [RRNet: Towards ReLU-Reduced Neural Network for Two-party Computation Based Private Inference](https://mlsanity.com/paper/2302.02292/) (2023)
- [PolyMPCNet: Towards ReLU-free Neural Architecture Search in Two-party Computation Based Private Inference](https://mlsanity.com/paper/2209.09424/) (2022)
- [Can persistent homology whiten Transformer-based black-box models? A case study on BERT compression](https://mlsanity.com/paper/2312.10702/) (2023)
- [EdgeFlex-Transformer: Transformer Inference for Edge Devices](https://mlsanity.com/paper/2512.19741/) (2025)
- [Efficient Traffic State Forecasting using Spatio-Temporal Network Dependencies: A Sparse Graph Neural Network Approach](https://mlsanity.com/paper/2211.03033/) (2022)
- [Block-wise Bit-Compression of Transformer-based Models](https://mlsanity.com/paper/2303.09184/) (2023)

## Abstract (arXiv)

With the yearning for deep learning democratization, there are increasing demands to implement Transformer-based natural language processing (NLP) models on resource-constrained devices for low-latency and high accuracy. Existing BERT pruning methods require domain experts to heuristically handcraft hyperparameters to strike a balance among model size, latency, and accuracy. In this work, we propose AE-BERT, an automatic and efficient BERT pruning framework with efficient evaluation to select a "good" sub-network candidate (with high accuracy) given the overall pruning ratio constraints. Our proposed method requires no human experts experience and achieves a better accuracy performance on many NLP tasks. Our experimental results on General Language Understanding Evaluation (GLUE) benchmark show that AE-BERT outperforms the state-of-the-art (SOTA) hand-crafted pruning methods on BERTBASE. On QNLI and RTE, we obtain 75% and 42.8% more overall pruning ratio while achieving higher accuracy. On MRPC, we obtain a 4.6 higher score than the SOTA at the same overall pruning ratio of 0.5. On STS-B, we can achieve a 40% higher pruning ratio with a very small loss in Spearman correlation compared to SOTA hand-crafted pruning methods. Experimental results also show that after model compression, the inference time of a single BERTBASE encoder on Xilinx Alveo U200 FPGA board has a 1.83× speedup compared to Intel(R) Xeon(R) Gold 5218 (2.30GHz) CPU, which shows the reasonableness of deploying the proposed method generated subnets of BERTBASE model on computation restricted devices.