Wenbin Li @ R&L Group

Wenbin Li (李文斌)
Associate Researcher (副研究员)

R&L Group
Department of Computer Science and Technology
State Key Laboratory for Novel Software Technology
Nanjing University

Email: [email protected]; [email protected]
Office: Room 1019, Computer Science Building, Xianlin Campus of Nanjing University, Nanjing 210023, China

[Google Scholar] [Github] [Github-VIG]

Biography

Currently, I am an Associate Researcher of Department of Computer Science and Technology at Nanjing University and a member of Reasoning and Learning Research Group, led by professor Yang Gao.

I received my Ph.D. degree in Department of Computer Science and Technology in December 2019 from Nanjing University.
I received my B.Sc. degree in School of Computer Science and Technology in June 2013 from China University of Mining and Technology.
I visited the VIStA group from December 2017 to June 2019 at the University of Rochester, under the supervision of Prof. Jiebo Luo.

News

[04/2024]: 课题组长期招收博士研究生和硕士研究生, 欢迎与我们联系!
[04/2024]: A Comprehensive Library for Continual Learning: LibContinual is released.
[05/2024]: One paper on “Imperfect Information Games” is accepted to ICML 2024.
[04/2024]: Two papers on “Activity Recognition” and “MARL” are accepted to IJCAI 2024.
[02/2024]: One paper on “Dataset Distillation” is accepted to CVPR 2024.
[01/2024]: One paper on “Neural Radiance Fields (NeRF)” is accepted to ICLR 2024.
[02/2024]: One paper on “Adversarial Training” is accepted to Neural Networks 2024.
[01/2024]: One paper on “Video Style Transfer” is accepted to CVIU 2024.
[12/2023]: One paper on “Multi-Agent Reinforcement Learning” is accepted to AAAI 2024.
[11/2023]: One paper on “Multi-Agent Reinforcement Learning” is accepted to TCYB 2023.
[09/2023]: One paper on “Subgame solving” is accepted to NeurIPS 2023.
[09/2023]: One paper on “Few-shot learning” is accepted to TPAMI 2023.
[07/2023]: One paper on “Neural Radiance Fields (NeRF)” is accepted to ACM MM 2023.
[07/2023]: One paper on “Out-of-distribution generalization” is accepted to ICCV 2023.
[05/2023]: To serve as Area Chair of Winter Conference on Applications of Computer Vision WACV 2024.
[05/2023]: One paper on “Self-supervised learning” is accepted to TMLR 2023.
[05/2023]: One paper on “Few-shot learning” is accepted to Pattern Recognition 2023.
[03/2023]: One paper on “Novel class discovery” is accepted to CVPR 2023.
[11/2022]: One paper on “Multi-agent reinforcement learning” is accepted to AAAI 2023.
[10/2022]: One paper on “Few-shot learning” is accepted to TPAMI 2023.
[10/2022]: One paper on “Online metric learning” is accepted to TNNLS 2022.
[09/2022]: Serve as Publication Chair of 16th Asian Conference on Computer Vision ACCV 2022.
[07/2022]: One paper on “Few-shot learning” is accepted to ECCV 2022.
[05/2022]: One paper on “Caricature generation” is accepted to TIP 2022.
[05/2022]: One paper on “Online learning” is accepted to TNNLS 2022.

Research Interests

My research interests include Machine Learning and Computer Vision. Especially in,

Few-shot Learning & Meta Learning
Machine Learning Compilation
Self-supervised Learning
Adversarial Learning
Continual Learning

Preprints

Dongdong Ren, Wenbin Li, Tianyu Ding, Lei Wang, Qi Fan, Jing Huo, Hongbing Pan and Yang Gao.
ONNXPruner: ONNX-Based General Model Pruning Adapter.
In: arXiv preprint arXiv:2404.08016, 2024.

Wenbin Li, Meihao Kong, Xuesong Yang, Lei Wang, Jing Huo, Yang Gao and Jiebo Luo.
A Unified Framework for Contrastive Learning from a Perspective of Affinity Matrix.
In: arXiv preprint arXiv:2211.14516, 2022.

Publications (Selected)

Journal Articles:

Changbin Shao, Wenbin Li*, Jing Huo, Zhenhua Feng, Yang Gao.
Attention-based Investigation and Solution to the Trade-off Issue of Adversarial Training.
In: Neural Networks, 2024.
(Impact Factor: 7.8)
[Paper] [Code]

Jing Huo, Meihao Kong, Wenbin Li, Jing Wu, Yu-Kun Lai, Yang Gao.
Towards Efficient Image and Video Style Transfer via Distillation and Learnable Feature Transformation.
In: Computer Vision and Image Understanding (CVIU), 2024.
(Impact Factor: 4.5)
[Paper] [Demo]

Wenbin Li, Ziyi Wang, Xuesong Yang, Chuanqi Dong, Pinzhuo Tian, Tiexin Qin, Jing Huo, Yinghuan Shi, Lei Wang, Yang Gao and Jiebo Luo.
LibFewShot: A Comprehensive Library for Few-shot Learning.
In: IEEE Transactions on Pattern Analysis and Machine Intelligence (TPAMI), 2023.
(Impact Factor: 23.6)
[Paper] [Code]

Wenbin Li, Lei Wang, Xingxing Zhang, Lei Qi, Jing Huo, Yang Gao, Jiebo Luo.
Defensive Few-shot Learning.
In: IEEE Transactions on Pattern Analysis and Machine Intelligence (TPAMI), 2023.
(Impact Factor: 23.6)
[Paper] [Code]

Wenbin Li, Xuesong Yang, Meihao Kong, Lei Wang, Jing Huo, Yang Gao and Jiebo Luo.
Trip-ROMA: Self-Supervised Learning with Triplets and Random Mappings.
In: Transactions on Machine Learning Research (TMLR), 2023.
[Paper] [Code]

Shaokang Dong, Hangyu Mao, Shangdong Yang, Shengyu Zhu, Wenbin Li*, Jianye Hao, Yang Gao
WToE: Learning When to Explore in Multiagent Reinforcement Learning.
In: IEEE Transactions on Cybernetics (TCYB), 2023.
(Impact Factor: 11.8)

Boyao Shi, Wenbin Li*, Jing Huo, Pengfei Zhu, Lei Wang, Yang Gao.
Global- and Local-aware Feature Augmentation with Semantic Orthogonality for Few-shot Image Classification.
In: Pattern Recognition, 2023.
(Impact Factor: 8.0)
[Paper] [Code]

Jing Huo, Xiangde Liu, Wenbin Li, Yang Gao, Hujun Yin, Jiebo Luo.
CAST: Learning Both Geometric and Texture Style Transfers for Effective Caricature Generation.
In: IEEE Transactions on Image Processing (TIP), 2022.
(Impact Factor: 10.6)
[Paper]

Wenbin Li, Yanfang Liu, Jing Huo, Yinghuan Shi, Yang Gao, Lei Wang, Jiebo Luo.
A Multilayer Framework for Online Metric Learning.
In: IEEE Transactions on Neural Networks and Learning Systems (TNNLS), 2022.
(Impact Factor: 10.4)

Yanfang Liu, Xiaocong Fan, Wenbin Li, Yang Gao.
Online Passive-Aggressive Active Learning for Trapezoidal Data Streams.
In: IEEE Transactions on Neural Networks and Learning Systems (TNNLS), 2022.
(Impact Factor: 10.4)
[Paper]

Pinzhuo Tian, Wenbin Li, Yang Gao.
Consistent Meta-Regularization for Better Meta-Knowledge in Few-Shot Learning.
In: IEEE Transactions on Neural Networks and Learning Systems (TNNLS), 2021.
(Impact Factor: 10.4)
[Paper]

Zheng Gu, Chuanqi Dong, Jing Huo, Wenbin Li, Yang Gao.
CariMe: Unpaired Caricature Generation with Multiple Exaggerations.
In: IEEE Transactions on Multimedia (TMM), 2021.
(Impact Factor: 7.3)
[Paper] [Code]

Wenbin Li, Wei Xiong, Haofu Liao, Jing Huo, Yang Gao, Jiebo Luo.
CariGAN: Caricature Generation through Weakly Paired Adversarial Learning.
In: Neural Networks, 2020.
(Impact Factor: 7.8)
[Paper]

Conference Articles:

Zhenxing Ge, Zheng Xu, Tianyu Ding, Linjian Meng, Bo An, Wenbin Li, Yang Gao.
Safe and Robust Subgame Exploitation in Imperfect Information Games.
In: International Conference on Machine Learning (ICML), 2024.
(Acceptance Rate: 2609/9473=27.5%)

Chao Li, Yujing Hu, Shangdong Yang, Tangjie Lv, Changjie Fan, Wenbin Li*, Chongjie Zhang, Yang Gao*.
STAR: Spatio-Temporal State Compression for Multi-Agent Tasks with Rich Observations.
In: International Joint Conference on Artificial Intelligence (IJCAI), 2024.

Rui Yan, Hongyu Qu, Xiangbo Shu, Wenbin Li*, Jinghui Tang, Tieniu Tan.
DTS-TPT: Dual Temporal-Sync Test-time Prompt Tuning for Zero-shot Activity Recognition.
In: International Joint Conference on Artificial Intelligence (IJCAI), 2024.

Wenxiao Deng, Wenbin Li*, Tianyu Ding, Lei Wang, Hongguang Zhang,
Kuihua Huang, Jing Huo, and Yang Gao.
Exploiting Inter-sample and Inter-feature Relations in Dataset Distillation.
In: IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2024.
(Acceptance Rate: 2719/11532=23.6%)
[arXiv] [Code]

Yanqi Bao, Tianyu Ding, Jing Huo, Wenbin Li, Yuxin Li, and Yang Gao.
InsertNeRF: Instilling Generalizability into NeRF with HyperNet Modules.
In: International Conference on Learning Representations (ICLR), 2024.
(Acceptance Rate: 2251/7262=31%)
[Paper] [Code]

Chao Li , Yupeng Zhang, Jianqi Wang, Yujing Hu, Shaokang Dong, Wenbin Li,
Tangjie Lv, Changjie Fan, Yang Gao.
Optimistic Value Instructors for Cooperative Multi-Agent Reinforcement Learning.
In: AAAI Conference on Artificial Intelligence (AAAI), 2024.
(Acceptance Rate: 2342/9862=23.75%)

Zhenxing Ge, Zheng Xu, Tianyu Ding, Wenbin Li*, Yang Gao*.
Efficient Subgame Refinement for Extensive-form Games.
In: Conference on Neural Information Processing Systems (NeurIPS), 2023.
(Acceptance Rate: 3222/12343=26.1%)

Yanqi Bao, Yuxin Li, Jing Huo, Tianyu Ding, Xinyue Liang, Wenbin Li, Yang Gao.
Where and How: Mitigating Confusion in Neural Radiance Fields from Sparse Inputs.
In: ACM Conference on Multimedia (ACM MM), 2023.
(Acceptance Rate: 902/3072=29.3%)

Min Zhang, Junkun Yuan, Yue He, Wenbin Li, Zhengyu Chen, Kun Kuang.
MAP: Towards Balanced Generalization of IID and OOD through Model-Agnostic Adapters.
In: International Conference on Computer Vision (ICCV), 2023. [Oral]

Wenbin Li, Zhichen Fan, Jing Huo, Yang Gao.
Modeling Inter-Class and Intra-Class Constraints in Novel Class Discovery.
In: IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2023.
(Acceptance Rate: 2360/9155=25.78%)
[Paper] [Code]

Wubing Chen, Wenbin Li*, Xiao Liu, Shangdong Yang, Yang Gao*.
Learning Explicit Credit Assignment for Cooperative Multi-Agent Reinforcement Learning
via Polarization Policy Gradient.
In: AAAI Conference on Artificial Intelligence (AAAI), 2023.
(Acceptance Rate: 1721/8777=19.6%)
[Paper] [Code]

Min Zhang, Siteng Huang, Wenbin Li, Donglin Wang.
Tree Structure-Aware Few-Shot Image Classification via Hierarchical Aggregation.
In: European Conference on Computer Vision (ECCV), 2022.
(Acceptance Rate: 1650/5803=28%)
[Paper] [Code]

Zheng Gu^#, Wenbin Li^#, Jing Huo, Lei Wang, Yang Gao.
LoFGAN: Fusing Local Representations for Few-shot Image Generation.
In: International Conference on Computer Vision (ICCV), 2021.
(Acceptance Rate: 1617/6236=25.9%) [# Equal contribution]
[Paper] [Code]

Jing Huo, Shiyin Jin, Wenbin Li, Jing Wu, Yu-Kun Lai, Yinghuan Shi, Yang Gao.
Manifold Alignment for Semantically Aligned Style Transfer.
In: International Conference on Computer Vision (ICCV), 2021. [Oral]
(Acceptance Rate: 1617/6236=25.9%; Oral: 210/6236=3%)
[Paper] [Code]

Wenbin Li, Lei Wang, Jing Huo, Yinghuan Shi, Yang Gao, Jiebo Luo.
Asymmetric Distribution Measure for Few-shot Learning.
In: International Joint Conference on Artificial Intelligence (IJCAI), 2020.
(Acceptance Rate: 592/4717=12.6%)
[Paper] [Code]

Chuanqi Dong, Wenbin Li, Jing Huo, Zheng Gu, Yang Gao.
Learning Task-aware Local Representations for Few-shot Learning.
In: International Joint Conference on Artificial Intelligence (IJCAI), 2020.
(Acceptance Rate: 592/4717=12.6%)
[Paper] [Code]

Jinglin Xu, Xiangsen Zhang, Wenbin Li, Xinwang Liu, Junwei Han.
Joint Multi-view 2D Convolutional Neural Networks for 3D Object Classification.
In: International Joint Conference on Artificial Intelligence (IJCAI), 2020.
(Acceptance Rate: 592/4717=12.6%)
[Paper]

Changbin Shao, Jing Huo, Lei Qi, Zhen-Hua Feng, Wenbin Li, Chuanqi Dong, Yang Gao.
Biased Feature Learning for Occlusion Invariant Face Recognition.
In: International Joint Conference on Artificial Intelligence (IJCAI), 2020.
(Acceptance Rate: 592/4717=12.6%)
[Paper]

Jinglin Xu, Wenbin Li, Xinwang Liu, Dingwen Zhang, Ji Liu, Junwei Han.
Deep Embedded Complementary and Interactive Information for Multi-view Classification.
In: AAAI Conference on Artificial Intelligence (AAAI), 2020.
(Acceptance Rate: 1591/7737=20.6%) [# Equal contribution]
[Paper] [Code]

Xiao Liu, Wenbin Li, Jing Huo, Lili Yao, Yang Gao.
Layerwise Sparse Coding for Pruned Deep Neural Networks with Extreme Compression Ratio.
In: AAAI Conference on Artificial Intelligence (AAAI), 2020.
(Acceptance Rate: 1591/7737=20.6%)
[Paper]

Wenbin Li, Lei Wang, Jinglin Xu, Jing Huo, Yang Gao, Jiebo Luo.
Revisiting Local Descriptor based Image-to-Class Measure for Few-shot Learning.
In: IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2019.
(Acceptance Rate: 1299/5165=25.2%)
[Paper] [Code]

Wenbin Li^#, Jinglin Xu^#, Jing Huo, Lei Wang, Yang Gao, Jiebo Luo.
Distribution Consistency based Covariance Metric Networks for Few-shot Learning.
In: AAAI Conference on Artificial Intelligence (AAAI), 2019. [Oral]
(Acceptance Rate: 1150/7095=16.2%) [# Equal contribution]
[Paper] [Slides] [Code]

Datasets & Framework

LibContinual

LibContinual is a comprehensive library for continual learning, especially for continual image classification. It integrates multiple classic continual methods into a unified framework. This library is friendly for beginners in continual learning with very concise code and clear structure..

LibFewShot

LibFewShot is a comprehensive library for few-shot learning, especially for few-shot image classification. It integrates multiple classic FSL methods into a unified framework, including four fine-tuning based methods, six meta-learning based methods, and eight metric-learning based methods.

WebCaricature Dataset

WebCaricature dataset is developed for caricature recognition. It is a large photograph-caricature dataset consisting of 6042 caricatures and 5974 photographs from 252 persons collected from the web.

Awards & Honors

2023 中国科协青年人才托举
2023 江苏省计算机学会青年科技奖
2023 南京大学本科教育教学杜夏奖教金
2021 江苏省双创博士
2021 南京大学紫金学者
2020 江苏省计算机学会优博
2020 中国研究生人工智能创新大赛优秀指导教师

Correspondence

Email: [email protected]; [email protected]
Office: Room 1019, Computer Science Building, Xianlin Campus of Nanjing University.
Address: Wenbin Li, State Key Laboratory for Novel Software Technology, Nanjing University, Xianlin Campus, 163 Xianlin Avenue,
Qixia District, Nanjing 210023, China
(南京市栖霞区仙林大道163号, 南京大学仙林校区, 软件新技术国家重点实验室, 210023.)