Preprint · 2020 · Open access

6.7ms on Mobile with over 78% ImageNet Accuracy: Unified Network Pruning and Architecture Search for Beyond Real-Time Mobile Acceleration.

Zhengang Li, Geng Yuan, Wei Niu, Yanyu Li, Pu Zhao, Yuxuan Cai, Xuan Shen, Zheng Zhan, Zhenglun Kong, Qing Jin, Zhiyu Chen, Sijia Liu and 4 more

arXiv (Cornell University) · Published 1 December 2020

Open full textCite this paper

In brief

“6.7ms on Mobile with over 78% ImageNet Accuracy: Unified Network Pruning and Architecture Search for Beyond Real-Time Mobile Acceleration.” is a 2020 preprint by Zhengang Li and 15 co-authors, published in arXiv (Cornell University). OpenAlex counts 1 citations as of 19 November 2026. It is open access: the full text is free to read at arXiv (Cornell University).

Citations
1
Most cited in
2021 (1)
References
23
Authors
16

Abstract

With the increasing demand to efficiently deploy DNNs on mobile edge devices, it becomes much more important to reduce unnecessary computation and increase the execution speed. Prior methods towards this goal, including model compression and network architecture search (NAS), are largely performed independently and do not fully consider compiler-level optimizations which is a must-do for mobile acceleration. In this work, we first propose (i) a general category of fine-grained structured pruning applicable to various DNN layers, and (ii) a comprehensive, compiler automatic code generation framework supporting different DNNs and different pruning schemes, which bridge the gap of model compression and NAS. We further propose NPAS, a compiler-aware unified network pruning, and architecture search. To deal with large search space, we propose a meta-modeling procedure based on reinforcement learning with fast evaluation and Bayesian optimization, ensuring the total number of training epochs comparable with representative NAS frameworks. Our framework achieves 6.7ms, 5.9ms, 3.9ms ImageNet inference times with 78.2%, 75% (MobileNet-V3 level), and 71% (MobileNet-V2 level) Top-1 accuracy respectively on an off-the-shelf mobile phone, consistently outperforming prior work.

Cite this paper

Formatted from the catalogue record. Check names, capitals and page numbers against the paper before you submit.

APA 7

Li, Z., Yuan, G., Niu, W., Li, Y., Zhao, P., Cai, Y., Shen, X., Zhan, Z., Kong, Z., Jin, Q., Chen, Z., Liu, S., Yang, K., Ren, B., Wang, Y., & Lin, X. (2020). 6.7ms on Mobile with over 78% ImageNet Accuracy: Unified Network Pruning and Architecture Search for Beyond Real-Time Mobile Acceleration. arXiv (Cornell University). https://arxiv.org/abs/2012.00596v1

MLA 9

Li, Zhengang, et al. “6.7ms on Mobile with over 78% ImageNet Accuracy: Unified Network Pruning and Architecture Search for Beyond Real-Time Mobile Acceleration.” arXiv (Cornell University), 2020. https://arxiv.org/abs/2012.00596v1.

Chicago

Li, Zhengang, Geng Yuan, Wei Niu, Yanyu Li, Pu Zhao, Yuxuan Cai, Xuan Shen, et al. 2020. “6.7ms on Mobile with over 78% ImageNet Accuracy: Unified Network Pruning and Architecture Search for Beyond Real-Time Mobile Acceleration.” arXiv (Cornell University). https://arxiv.org/abs/2012.00596v1.

Harvard

Li, Z. et al. (2020) ‘6.7ms on Mobile with over 78% ImageNet Accuracy: Unified Network Pruning and Architecture Search for Beyond Real-Time Mobile Acceleration.’, arXiv (Cornell University). Available at: https://arxiv.org/abs/2012.00596v1.

IEEE

Z. Li et al., “6.7ms on Mobile with over 78% ImageNet Accuracy: Unified Network Pruning and Architecture Search for Beyond Real-Time Mobile Acceleration.,” arXiv (Cornell University), 2020.

BibTeX

@misc{li2020mobile,
  title     = {{6.7ms on Mobile with over 78\% ImageNet Accuracy: Unified Network Pruning and Architecture Search for Beyond Real-Time Mobile Acceleration.}},
  author    = {Li, Zhengang and Yuan, Geng and Niu, Wei and Li, Yanyu and Zhao, Pu and Cai, Yuxuan and Shen, Xuan and Zhan, Zheng and Kong, Zhenglun and Jin, Qing and Chen, Zhiyu and Liu, Sijia and Yang, Kaiyuan and Ren, Bin and Wang, Yanzhi and Lin, Xue},
  howpublished = {arXiv (Cornell University)},
  year      = {2020},
  url       = {https://arxiv.org/abs/2012.00596v1}
}

Download for a reference manager: BibTeX (.bib) · RIS for Zotero, Mendeley and EndNote (.ris)

Topic and keywords

Physical Sciences › Computer Science › Computer Vision and Pattern Recognition › Advanced Neural Network Applications

  • network pruning
  • neural architecture search
  • mobile edge devices
  • automatic code generation
  • metamodeling
  • reinforcement learning
  • Bayesian optimization
  • inference speed

Work with this paper

Most-cited papers in Computer Science

All highly cited Computer Science papers

Questions about this paper

How do I cite this paper in APA style?

Li, Z., Yuan, G., Niu, W., Li, Y., Zhao, P., Cai, Y., Shen, X., Zhan, Z., Kong, Z., Jin, Q., Chen, Z., Liu, S., Yang, K., Ren, B., Wang, Y., & Lin, X. (2020). 6.7ms on Mobile with over 78% ImageNet Accuracy: Unified Network Pruning and Architecture Search for Beyond Real-Time Mobile Acceleration. arXiv (Cornell University). https://arxiv.org/abs/2012.00596v1

How many times has this paper been cited?

1 times, according to OpenAlex on 19 November 2026. Google Scholar and Scopus count different sets of sources, so their numbers differ.

Is this paper free to read?

Yes. It is open access at arXiv (Cornell University): sign in and open the full text from this page.

Who wrote this paper?

Zhengang Li, Geng Yuan, Wei Niu, Yanyu Li, Pu Zhao, Yuxuan Cai, Xuan Shen, Zheng Zhan, Zhenglun Kong, Qing Jin, Zhiyu Chen, Sijia Liu and 4 more.

Bibliographic data from OpenAlex (CC0), last checked 19 November 2026. ResearchGuru does not host this paper; once you are signed in, the links open it at the publisher or the repository that does.