Picture for Ao Liu

Ao Liu

FedCliP: Federated Learning with Client Pruning

Jan 17, 2023
Beibei Li, Zerui Shao, Ao Liu, Peiran Wang

Figure 1 for FedCliP: Federated Learning with Client Pruning

Figure 2 for FedCliP: Federated Learning with Client Pruning

Figure 3 for FedCliP: Federated Learning with Client Pruning

* 8 pages, 1 figure, 2 tables

Access Paper or Ask Questions

Exploiting Unlabeled Data for Target-Oriented Opinion Words Extraction

Aug 17, 2022
Yidong Wang, Hao Wu, Ao Liu, Wenxin Hou, Zhen Wu, Jindong Wang, Takahiro Shinozaki, Manabu Okumura, Yue Zhang

Figure 1 for Exploiting Unlabeled Data for Target-Oriented Opinion Words Extraction

Figure 2 for Exploiting Unlabeled Data for Target-Oriented Opinion Words Extraction

Figure 3 for Exploiting Unlabeled Data for Target-Oriented Opinion Words Extraction

Figure 4 for Exploiting Unlabeled Data for Target-Oriented Opinion Words Extraction

* Accepted by COLING 2022

Access Paper or Ask Questions

A Unified Weight Initialization Paradigm for Tensorial Convolutional Neural Networks

May 28, 2022
Yu Pan, Zeyong Su, Ao Liu, Jingquan Wang, Nannan Li, Zenglin Xu

Figure 1 for A Unified Weight Initialization Paradigm for Tensorial Convolutional Neural Networks

Figure 2 for A Unified Weight Initialization Paradigm for Tensorial Convolutional Neural Networks

Figure 3 for A Unified Weight Initialization Paradigm for Tensorial Convolutional Neural Networks

Figure 4 for A Unified Weight Initialization Paradigm for Tensorial Convolutional Neural Networks

* Accepted in ICML 2022

Access Paper or Ask Questions

Semi-Supervised Formality Style Transfer with Consistency Training

Mar 25, 2022
Ao Liu, An Wang, Naoaki Okazaki

Figure 1 for Semi-Supervised Formality Style Transfer with Consistency Training

Figure 2 for Semi-Supervised Formality Style Transfer with Consistency Training

Figure 3 for Semi-Supervised Formality Style Transfer with Consistency Training

Figure 4 for Semi-Supervised Formality Style Transfer with Consistency Training

* ACL 2022 main conference

Access Paper or Ask Questions