Qiushi Li 李秋实

Selected Publications

systems for efficiency and security

• Manifold Projection: A Unified View to Break and Fix Weight Obfuscation in TEE-Shielded LLM Partition
  Jingwen Huang, Qiushi Li*, Yezheng Cheng, Ju Ren
  IEEE Symposium on Security and Privacy (IEEE S&P, to appear)

• Nappa: NNA-Compatible and Privacy-Preserving DNN Training Framework via Vector Decomposition [paper]
  Yan Zhang, Qiushi Li*, Ju Ren, Yiqiao Liao, Jin Ouyang, Chengru Song, Honghuan Wu, Kaiqiao Zhan, Ben Wang, Xu Chen, Yaoxue Zhang
  IEEE Transactions on Dependable and Secure Computing (TDSC)

• Privacy-Preserving DNN Training with Prefetched Meta-Keys on Heterogeneous Neural Network Accelerators [paper]
  Qiushi Li, Ju Ren, Yan Zhang, Chengru Song, Yiqiao Liao, Yaoxue Zhang
  IEEE/ACM Design Automation Conference (DAC)

• ENIGMA: Low-Latency and Privacy-Preserving Edge Inference on Heterogeneous Neural Network Accelerators [paper]
  Qiushi Li, Ju Ren, Xinglin Pan, Yuezhi Zhou, Yaoxue Zhang
  IEEE International Conference on Distributed Computing Systems (ICDCS)

• Privacy-Preserving DNN Model Authorization against Model Theft and Feature Leakage [paper]
  Qiushi Li, Ju Ren, Yuezhi Zhou, Yaoxue Zhang
  IEEE International Conference on Communications (ICC)

privacy boundaries

• User-Defined Visual Privacy Preserving via Vision–Language Region Anonymization
  Yan Zhang, Junfeng Fang, Zhenya Ma, Manzhou Li, Qiushi Li*, Ju Ren
  IEEE International Conference on Multimedia and Expo (ICME, to appear)

• CSVAR: Enhancing Visual Privacy in Federated Learning via Adaptive Shuffling Against Overfitting [paper]
  Ye Guo, Zhenya Ma, Yan Zhang, Donghua Cai, Qiushi Li*, Yongheng Deng, Ye Zhang, Yaxi Liu, Boyang Zhou, Ju Ren, Xuemin Shen
  IEEE International Conference on Communications (ICC)

• PeriNet: Periodic Deep Learning Framework for Modality-Agnostic Privacy Preserving [paper]
  Yan Zhang, Yihong Song, Manzhou Li, Qiushi Li*, Qi Li and Ju Ren
  ACM on Web Conference (WWW)

• You Can Use But Cannot Recognize: Preserving Visual Privacy in Deep Neural Networks [paper]
  Qiushi Li, Yan Zhang, Ju Ren, Qi Li, Yaoxue Zhang
  Network and Distributed System Security (NDSS) Symposium

• InvisibleFL: Federated Learning over Non-Informative Intermediate Updates against Multimedia Privacy Leakages [paper]
  Qiushi Li, Wenwu Zhu, Chao Wu, Xinglin Pan, Fan Yang, Yuezhi Zhou, Yaoxue Zhang
  ACM International Conference on Multimedia (MM)

* co-first or corresponding author.
For more publications, see here.