Get our free extension to see links to code for papers anywhere online!

Add to Chrome

Add to Firefox

Picture for Junyi Li

Junyi Li

A Fully Single Loop Algorithm for Bilevel Optimization without Hessian Inverse

Dec 10, 2021
Junyi Li, Bin Gu, Heng Huang

Figure 1 for A Fully Single Loop Algorithm for Bilevel Optimization without Hessian Inverse

Figure 2 for A Fully Single Loop Algorithm for Bilevel Optimization without Hessian Inverse

Figure 3 for A Fully Single Loop Algorithm for Bilevel Optimization without Hessian Inverse

Figure 4 for A Fully Single Loop Algorithm for Bilevel Optimization without Hessian Inverse

* To appear in AAAI 2022

Access Paper or Ask Questions

SUPER-ADAM: Faster and Universal Framework of Adaptive Gradients

Jun 30, 2021
Feihu Huang, Junyi Li, Heng Huang

Figure 1 for SUPER-ADAM: Faster and Universal Framework of Adaptive Gradients

Figure 2 for SUPER-ADAM: Faster and Universal Framework of Adaptive Gradients

Figure 3 for SUPER-ADAM: Faster and Universal Framework of Adaptive Gradients

* 18 pages, 5 figures. We add the detailed proofs and correct some typos

Access Paper or Ask Questions

Compositional Federated Learning: Applications in Distributionally Robust Averaging and Meta Learning

Jun 21, 2021
Feihu Huang, Junyi Li, Heng Huang

Figure 1 for Compositional Federated Learning: Applications in Distributionally Robust Averaging and Meta Learning

Figure 2 for Compositional Federated Learning: Applications in Distributionally Robust Averaging and Meta Learning

Figure 3 for Compositional Federated Learning: Applications in Distributionally Robust Averaging and Meta Learning

Figure 4 for Compositional Federated Learning: Applications in Distributionally Robust Averaging and Meta Learning

* 21 pages, 8 figures

Access Paper or Ask Questions

Few-shot Knowledge Graph-to-Text Generation with Pretrained Language Models

Jun 03, 2021
Junyi Li, Tianyi Tang, Wayne Xin Zhao, Zhicheng Wei, Nicholas Jing Yuan, Ji-Rong Wen

Figure 1 for Few-shot Knowledge Graph-to-Text Generation with Pretrained Language Models

Figure 2 for Few-shot Knowledge Graph-to-Text Generation with Pretrained Language Models

Figure 3 for Few-shot Knowledge Graph-to-Text Generation with Pretrained Language Models

Figure 4 for Few-shot Knowledge Graph-to-Text Generation with Pretrained Language Models

* Accepted to ACL 2021 Findings

Access Paper or Ask Questions

Pretrained Language Models for Text Generation: A Survey

May 25, 2021
Junyi Li, Tianyi Tang, Wayne Xin Zhao, Ji-Rong Wen

Figure 1 for Pretrained Language Models for Text Generation: A Survey

Figure 2 for Pretrained Language Models for Text Generation: A Survey

* Accepted by IJCAI 2021 Survey Track

Access Paper or Ask Questions

Knowledge-based Review Generation by Coherence Enhanced Text Planning

May 09, 2021
Junyi Li, Wayne Xin Zhao, Zhicheng Wei, Nicholas Jing Yuan, Ji-Rong Wen

Figure 1 for Knowledge-based Review Generation by Coherence Enhanced Text Planning

Figure 2 for Knowledge-based Review Generation by Coherence Enhanced Text Planning

Figure 3 for Knowledge-based Review Generation by Coherence Enhanced Text Planning

Figure 4 for Knowledge-based Review Generation by Coherence Enhanced Text Planning

* Accepted by SIGIR 2021 (Long Paper)

Access Paper or Ask Questions

WenLan: Bridging Vision and Language by Large-Scale Multi-Modal Pre-Training

Mar 19, 2021
Yuqi Huo, Manli Zhang, Guangzhen Liu, Haoyu Lu, Yizhao Gao, Guoxing Yang, Jingyuan Wen, Heng Zhang, Baogui Xu, Weihao Zheng, Zongzheng Xi, Yueqian Yang, Anwen Hu, Jinming Zhao, Ruichen Li, Yida Zhao, Liang Zhang, Yuqing Song, Xin Hong, Wanqing Cui, Danyang Hou, Yingyan Li, Junyi Li, Peiyu Liu, Zheng Gong, Chuhao Jin, Yuchong Sun, Shizhe Chen, Zhiwu Lu, Zhicheng Dou, Qin Jin, Yanyan Lan, Wayne Xin Zhao, Ruihua Song, Ji-Rong Wen

Figure 1 for WenLan: Bridging Vision and Language by Large-Scale Multi-Modal Pre-Training

Figure 2 for WenLan: Bridging Vision and Language by Large-Scale Multi-Modal Pre-Training

Figure 3 for WenLan: Bridging Vision and Language by Large-Scale Multi-Modal Pre-Training

Figure 4 for WenLan: Bridging Vision and Language by Large-Scale Multi-Modal Pre-Training

* This paper is the outcome of the Chinese multi-modal pre-training project called 'WenLan'

Access Paper or Ask Questions