Theses
Constrained codes and signal processing for non-volatile memories, Ph.D Thesis, University of California, San Diego, September 2014
Human gender recognition based on face features, B.E. Thesis, Tsinghua University, Beijing, China, July 2009
Publications
DA3D: Domain-aware dynamic adaptation for all-weather multimodal 3D detection, ACM International Conference on Multimedia, 2025
A min-Max optimization framework for sparse multi-task deep neural networks, Neurocomputing, Vol. 650, 2025
A unified DNN weight compression framework using reweighted optimization methods, Journal of Intelligent Systems with Applications, Vol. 27, 2025
Robust multi-task adversarial attacks using min-max optimization, IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP) 2025
An efficient and accurate dynamic sparse training framework based on parameter freezing, Association for the Advancement of Artificial Intelligence (AAAI) 2025
Data overfitting for on-device super-resolution with dynamic algorithm and compiler co-design, European Conference on Computer Vision (ECCV), 2024
Advancing dynamic sparse training by exploring optimization opportunities, International Conference on Machine Learning (ICML), 2024
NeurRev: Train better sparse neural network practically via neuron revitalization, International Conference on Learning Representations (ICLR) 2024
DISCO: Distributed inference with sparse communications, IEEE Winter Conference on Applications of Computer Vision (WACV), 2024
Condense: A framework for device and frequency adaptive neural network models on the edge, Design Automation Conference (DAC), 2023
Data level lottery ticket hypothesis for vision transformers, International Joint Conference on Artificial Intelligence (IJCAI) 2023
Towards high-quality and efficient video super-resolution via spatial-temporal data overfitting, IEEE Computer Vision and Pattern Recognition (CVPR), 2023
Pruning parameterization with bi-level optimization for efficient semantic segmentation on the edge, IEEE Computer Vision and Pattern Recognition (CVPR), 2023
Self-ensemble protection: Training checkpoints are good data protectors, International Conference on Learning Representations (ICLR) 2023
Towards real-time segmentation on the edge, Association for the Advancement of Artificial Intelligence (AAAI) 2023
Peeling the onion: hierarchical reduction of data redundancy for efficient vision transformer training, Association for the Advancement of Artificial Intelligence (AAAI) 2023
ESRU: Extremely low-bit and hardware-efficient stochastic rounding unit design for low-bit DNN training, Design, Automation & Test in Europe Conference & Exhibition (DATE) 2023
Compiler-aware neural architecture search for on-mobile real-time super-resolution, European Conference on Computer Vision (ECCV), 2022
SPViT: Enabling faster vision transformers via soft token pruning, European Conference on Computer Vision (ECCV), 2022
You Already Have It: A generator-free low-precision DNN training framework using stochastic rounding, European Conference on Computer Vision (ECCV), 2022
All-in-One: A highly representative DNN pruning framework for edge devices with dynamic power management, IEEE/ACM International Conference on Computer-Aided Design (ICCAD), 2022
CHEX: CHannel EXploration for CNN model compression, IEEE Computer Vision and Pattern Recognition (CVPR), 2022
Shfl-BW: Accelerating deep neural network inference with tensor-core aware weight pruning, Design Automation Conference (DAC), 2022
Hardware-efficient stochastic rounding unit design for DNN training: late breaking results, Design Automation Conference (DAC), 2022
Effective model sparsification by scheduled grow-and-prune methods, International Conference on Learning Representations (ICLR), 2022
Learning from the CNN-based compressed domain, IEEE Winter Conference on Applications of Computer Vision (WACV), 2022
Sanity checks for lottery tickets: Does your winning ticket really win the jackpot?, Neural Information Processing Systems (NeurIPS), 2021
MEST: Accurate and fast memory-economic sparse training framework on the edge, Neural Information Processing Systems (NeurIPS), 2021
Learning in the frequency domain, IEEE Computer Vision and Pattern Recognition (CVPR), 2020
Computation on sparse neural networks: an inspiration for future hardware, Design Automation Conference (DAC), 2020
Garbage collection algorithms meta data updates in NAND Flash, IEEE International Conference on Communications (ICC), 2019
A binarized neural network accelerator with differential crosspoint memristor array for energy-efficient Mac operations, International Symposiums on Circuits and Systems (ISCAS), 2019
Non-volatile memory array based quantization- and noise-resilient LSTM neural networks, IEEE International Conference on Rebooting Computing (ICRC), 2019
Improving noise tolerance of hardware accelerated artificial neural networks, IEEE International Conference on Machine Learning and Applications (ICMLA), 2018
Training recurrent neural networks against noisy computations during inference, Asilomar Conference on Signals, Systems and Computers, 2018
Improving robustness of neural networks against bit flipping errors during inference, International Conference on Image and Video Processing, 2018
Hamming distance based binary representation of numbers, IEEE International Symposium on Information Theory (ISIT), 2018
Improving robustness of neural networks against bit flipping errors during inference, Journal of Images and Graphics, 2018
Fractional bits-per-cell for NAND flash with low read latency, IEEE Global Communications Conference (Globecom), 2017
Enhancing the expected lifetime of NAND flash by short q-ary WOM codes, IEEE Communications Letters, vol. 22, no. 7, pp. 1302 - 1305, April 2018
Low read latency rewriting codes for multi-level 3D NAND flash, IEEE Communications Letters, vol. 21, no. 7, pp. 1477 - 1480, April 2017
Polar code constructions based on LLR evolution, IEEE Communications Letters, vol. 21, no. 6, pp. 1221 - 1224, January 2017
Joint source-channel decoding of polar codes for language-based sources, IEEE Global Communications Conference (Globecom), 2016
An alternative proof of channel polarization for channels with arbitrary input alphabets, Allerton Conference on Communication, Control, and Computing, 2015
Balanced codes for data retention of multi-level flash memories with fast page read, Allerton Conference on Communication, Control, and Computing, 2015
Adaptive read thresholds for NAND flash, IEEE Transactions on Communications, vol. 63, no. 9, pp. 3069 - 3081, July 2015
Enhanced belief propagation decoding of polar codes through concatenation, IEEE International Symposium on Information Theory (ISIT), 2014
Optimized cell programming for flash memories with quantizers, IEEE Transactions on Information Theory, vol. 60, no. 5, pp. 1-16, May 2014
Constrained codes that mitigate inter-cell interference in read/write cycles for flash memories, IEEE Journal on Selected Areas in Communications, vol. 32, no. 5, pp. 836-846, May 2014
Lattice-based WOM codes for multilevel flash memories, IEEE Journal on Selected Areas in Communications, vol. 32, no. 5, pp. 933-945, May 2014
Time-space constrained codes for phase-change memories, IEEE Transactions on Information Theory, vol. 59, no. 8, pp. 5102-5114, August 2013
Parallel programming of rank modulation, IEEE International Symposium on Information Theory (ISIT), 2013
Towards minimizing read time for NAND Flash, IEEE Global Communications Conference (Globecom), 2012
Optimized cell programming for flash memories with quantizers, IEEE International Symposium on Information Theory (ISIT), 2012
WOM with retained messages, IEEE International Symposium on Information Theory (ISIT), 2012
Time-space constrained codes for phase-change memories, IEEE Global Communications Conference (Globecom), 2011
A novel HVS-based SSIM on video quality assessment, China Communications Technology, 2010 (02) pp.85-87+90
A novel motion estimation method based on HVS-based similarity for inter prediction in H.264/AVC, China Communications Technology, 2009 (05) pp.103-105
Preprint
SPViT: Enabling Faster Vision Transformers via Soft Token Pruning, arXiv:2112.13890, December 2021
Load-balanced Gather-scatter Patterns for Sparse Deep Neural Networks, arXiv:2112.10898, December 2021
Sanity Checks for Lottery Tickets: Does Your Winning Ticket Really Win the Jackpot? arXiv:2107.00166, July 2021
Effective Model Sparsification by Scheduled Grow-and-Prune Methods, arXiv:2106.09857, June 2021
A unified DNN weight compression framework using reweighted optimization methods, arXiv:2004.05531, April 2020
Computation on sparse neural networks: an inspiration for future hardware, arXiv:2004.11946, April 2020
Noisy Computations during Inference: Harmful or Helpful?, arXiv:1811.10649, November, 2018
Robustness of neural networks against storage media errors, arXiv:1709.06173, September, 2017
Workshops
Robustness of neural networks against storage media errors, Information Theory and Application (ITA) workshop 2018, La Jolla, CA, USA, February 2018
Hamming distance based binary representation of numbers, 9th Non-volatile memories workshop, La Jolla, CA, USA, March 2018
Static garbage collection algorithms for meta data updates in NAND flash, 8th Non-volatile memories workshop, La Jolla, CA, USA, March 2017
High speed soft decision decoding architecture for triple level cell NAND flash memory, 7th Non-volatile memories workshop, La Jolla, CA, USA, March 2016
Enhancing the average lifetime of flash memory by lattice-based WOM codes, 6th Non-volatile memories workshop, La Jolla, CA, USA, March 2015
Coding and signal processing for non-volatile memories, Information Theory and Application Workshop (ITA), San Diego, CA, USA, February, 2014
Inter-cell interference free codes for read/write in flash memories, 4th Non-volatile memories workshop, La Jolla, CA, USA, March 2013
WOM with history, 3rd Non-volatile memories workshop, La Jolla, CA, USA, March 2012
Time-space constrained codes for phase-change memories, Flash Memory Summit 2011, Santa Clara, CA, USA, August 2011
Sum-capacity of multiple-write noisy memories, 2nd Non-volatile memories workshop, La Jolla, CA, USA, March 2011