Biography Publications Preprints Datasets Students Teaching Awards Services

Wenbin Li, PhD  

Associate Professor

State Key Laboratory for Novel Software Technology
School of Intelligence Science and Technology
Reasoning and Learning Research Group
Nanjing University, Suzhou Campus, China

Email: liwenbin@nju.edu.cn; liwenbin.nju@gmail.com

Wenbin Li

Biography

I am currently a Tenure-Track Associate Professor of School of Intelligence Science and Technology at Nanjing University and a member of Reasoning and Learning Research Group, led by Prof. Yang Gao. I received my Ph.D. degree in School of Computer Science in December 2019 from Nanjing University. I visited the VIStA group from December 2017 to June 2019 at the University of Rochester, under the supervision of Prof. Jiebo Luo.

Research Interests

My research interests encompass,

  • Brain-inspired Artificial Intelligence: I am particularly focused on few-shot concept learning, continual learning, and brain-inspired cognitive computation models, aiming to advance human-level cognitive intelligence.
  • Efficient Generative AI System: My attention is directed towards hardware-software codesign, model compression, LLM quantization, machine learning compilation, and their practical applications in generative AI models for robotics.
  • Multimodal Foundation Models: In addition, I also have a broad interest in multimodal foundation models, model merging, multi-modal reasoning, as well as the deployment and acceleration of large-scale models on edge devices.

    • News

    Selected Journal Publications

    ( # indicates equal contribution, * indicates corresponding authorship)
    1. ONNXPruner: ONNX-Based General Model Pruning Adapter
      Dongdong Ren, Wenbin Li*, Tianyu Ding, Lei Wang, Qi Fan, Jing Huo, Hongbing Pan, and Yang Gao.
      IEEE Transactions on Pattern Analysis and Machine Intelligence (TPAMI), 2025. (IF = 18.6)
    2. Leveraging Frequency Analysis for Image Denoising Network Pruning
      Dongdong Ren, Wenbin Li*, Tianyu Ding, Jing Huo, Lei Wang, Hongbing Pan, and Yang Gao.
      IEEE Transactions on Image Processing (TIP), 2025. (IF = 13.7)
    3. LibFewShot: A Comprehensive Library for Few-shot Learning
      Wenbin Li, Ziyi Wang, Xuesong Yang, Chuanqi Dong, Pinzhuo Tian, Tiexin Qin, Jing Huo, Yinghuan Shi, Lei Wang, Yang Gao and Jiebo Luo.
      IEEE Transactions on Pattern Analysis and Machine Intelligence (TPAMI), 2023. (IF = 18.6)
    4. Defensive Few-shot Learning
      Wenbin Li, Lei Wang, Xingxing Zhang, Lei Qi, Jing Huo, Yang Gao, Jiebo Luo.
      IEEE Transactions on Pattern Analysis and Machine Intelligence (TPAMI), 2023. (IF = 18.6)
    5. Trip-ROMA: Self-Supervised Learning with Triplets and Random Mappings
      Wenbin Li, Xuesong Yang, Meihao Kong, Lei Wang, Jing Huo, Yang Gao and Jiebo Luo.
      Transactions on Machine Learning Research (TMLR), 2023
    6. A Multilayer Framework for Online Metric Learning
      Wenbin Li, Yanfang Liu, Jing Huo, Yinghuan Shi, Yang Gao, Lei Wang, Jiebo Luo.
      IEEE Transactions on Neural Networks and Learning Systems (TNNLS), 2023. (IF = 8.9)

    Selected Conference Publications

    ( # indicates equal contribution, * indicates corresponding authorship)
    1. Branch, or Layer? Zeroth-Order Optimization for Continual Learning of Vision-Language Models
      Ziwei Liu#, Borui Kang#, Wei Li, Hangjie Yuan, Yanbing Yang, Wenbin Li, Jun Luo, Yifan Zhu, Tao Feng The 40th Annual AAAI Conference on Artificial Intelligence (AAAI), 2026. (AC Rate = 17.6%)
    2. Last-Iterate Convergence of Smooth Regret Matching+ Variants in Learning Nash Equilibria
      Linjian Meng, Youzhi Zhang, Zhenxing Ge, Tianyu Ding, Shangdong Yang, Zheng Xu, Wenbin Li, Yang Gao.
      The Annual Conference on Neural Information Processing Systems (NeurIPS), 2025. (AC Rate = 24.52%)
    3. Parameter-Free Last-Iterate Convergence of Counterfactual Regret Minimization Algorithms
      Linjian Meng, Youzhi Zhang, Shangdong Yang, Tianyu Ding, Zhenxing Ge, Wenbin Li, Tianpei Yang, Bo An, Yang Gao.
      The Annual Conference on Neural Information Processing Systems (NeurIPS), 2025. (AC Rate = 24.52%)
    4. Don’t Need Retraining: A Mixture of DETR and Vision Foundation Models for Cross-Domain Few-Shot Object Detection
      Chang-Han Liu, Xunzhi Xiang, Zixuan Duan, Wenbin Li, Yang Gao, Qi Fan.
      The Annual Conference on Neural Information Processing Systems (NeurIPS), 2025. (AC Rate = 24.52%)
    5. AT-Drone: Benchmarking Adaptive Teaming in Multi-Drone Pursuit
      Yang Li, Junfan Chen, Feng Xue, Jiabin Qiu, Wenbin Li, Qingrui Zhang, Ying Wen, Wei Pan. The Conference on Robot Learning (CoRL), 2025. (AC Rate = 35.77%)
    6. Dynamic Multi-Layer Null Space Projection for Vision-Language Continual Learning
      Borui Kang, Lei Wang, Zhiping Wu, Tao Feng, Yawen Li, Yang Gao, Wenbin Li*.
      International Conference on Computer Vision (ICCV), 2025. (AC Rate = 24.0%)
    7. Adapting In-Domain Few-Shot Segmentation to New Domains without Retraining
      Qi Fan, Kai-Qi Liu, Nian Liu, Hisham Cholakkal, Rao Muhammad Anwer, Wenbin Li, Yang Gao.
      International Conference on Computer Vision (ICCV), 2025. (AC Rate = 24.0%)
    8. Reducing Variance of Stochastic Optimization for Approximating Nash Equilibria in Normal-Form Games
      Linjian Meng, Youzhi Zhang, Wubing Chen, Wenbin Li, Tianpei Yang, and Yang Gao.
      International Conference on Machine Learning (ICML), 2025. [Spotlight poster] (AC Rate = 2.6%)
    9. Enhancing Few-Shot Class-Incremental Learning via Training-Free Bi-Level Modality Calibration
      Yiyang Chen, Tianyu Ding, Lei Wang, Jing Huo, Yang Gao, and Wenbin Li*.
      Computer Vision and Pattern Recognition (CVPR), 2025. (AC Rate = 22.1%)
    10. Exploiting Inter-sample and Inter-feature Relations in Dataset Distillation
      Wenxiao Deng, Wenbin Li*, Tianyu Ding, Lei Wang, Hongguang Zhang, Kuihua Huang, Jing Huo, and Yang Gao.
      Computer Vision and Pattern Recognition (CVPR), 2024. (AC Rate = 23.6%)
    11. Safe and Robust Subgame Exploitation in Imperfect Information Games
      Zhenxing Ge, Zheng Xu, Tianyu Ding, Linjian Meng, Bo An, Wenbin Li, and Yang Gao.
      International Conference on Machine Learning (ICML), 2024. (AC Rate = 27.5%)
    12. STAR: Spatio-Temporal State Compression for Multi-Agent Tasks with Rich Observations
      Chao Li, Yujing Hu, Shangdong Yang, Tangjie Lv, Changjie Fan, Wenbin Li*, Chongjie Zhang, Yang Gao*.
      International Joint Conference on Artificial Intelligence (IJCAI), 2024. (AC Rate = 14.3%)
    13. DTS-TPT: Dual Temporal-Sync Test-time Prompt Tuning for Zero-shot Activity Recognition
      Rui Yan, Hongyu Qu, Xiangbo Shu, Wenbin Li*, Jinghui Tang, Tieniu Tan.
      International Joint Conference on Artificial Intelligence (IJCAI), 2024. (AC Rate = 14.3%)
    14. InsertNeRF: Instilling Generalizability into NeRF with HyperNet Modules
      Yanqi Bao, Tianyu Ding, Jing Huo, Wenbin Li, Yuxin Li, and Yang Gao.
      International Conference on Learning Representations (ICLR), 2024. (AC Rate = 30.96%)
    15. Optimistic Value Instructors for Cooperative Multi-Agent Reinforcement Learning
      Chao Li , Yupeng Zhang, Jianqi Wang, Yujing Hu, Shaokang Dong, Wenbin Li,
      Tangjie Lv, Changjie Fan, Yang Gao.
      AAAI Conference on Artificial Intelligence (AAAI), 2024. (AC Rate = 23.75%)
    16. Efficient Subgame Refinement for Extensive-form Games
      Zhenxing Ge, Zheng Xu, Tianyu Ding, Wenbin Li*, and Yang Gao*.
      Neural Information Processing Systems (NeurIPS), 2023. (AC Rate = 26.1%)
    17. Where and How: Mitigating Confusion in Neural Radiance Fields from Sparse Inputs
      Yanqi Bao, Yuxin Li, Jing Huo, Tianyu Ding, Xinyue Liang, Wenbin Li, and Yang Gao.
      ACM International Conference on Multimedia (ACM MM), 2023. (AC Rate = 29.3%)
    18. MAP: Towards Balanced Generalization of IID and OOD through Model-Agnostic Adapters
      Min Zhang, Junkun Yuan, Yue He, Wenbin Li, Zhengyu Chen, Kun Kuang.
      International Conference on Computer Vision (ICCV), 2023. (AC Rate = 25.0%) [Oral]
    19. Modeling Inter-Class and Intra-Class Constraints in Novel Class Discovery
      Wenbin Li, Zhichen Fan, Jing Huo, Yang Gao.
      IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2023. (AC Rate = 25.8%)
    20. Learning Explicit Credit Assignment for Cooperative Multi-Agent Reinforcement Learning
      Wubing Chen, Wenbin Li*, Xiao Liu, Shangdong Yang, Yang Gao.*
      AAAI Conference on Artificial Intelligence (AAAI), 2023. (AC Rate = 19.6%) [Oral]
    21. Tree Structure-Aware Few-Shot Image Classification via Hierarchical Aggregation
      Min Zhang, Siteng Huang, Wenbin Li, Donglin Wang.
      European Conference on Computer Vision (ECCV), 2022. (AC Rate = 28.4%)
    22. LoFGAN: Fusing Local Representations for Few-shot Image Generation
      Zheng Gu#, Wenbin Li#, Jing Huo, Lei Wang, Yang Gao.
      International Conference on Computer Vision (ICCV), 2021. (AC Rate = 26.2%)
    23. Manifold Alignment for Semantically Aligned Style Transfer
      Jing Huo, Shiyin Jin, Wenbin Li*, Jing Wu, Yu-Kun Lai, Yinghuan Shi, Yang Gao.
      International Conference on Computer Vision (ICCV), 2021. (AC Rate = 26.2%)
    24. Asymmetric Distribution Measure for Few-shot Learning
      Wenbin Li, Lei Wang, Jing Huo, Yinghuan Shi, Yang Gao, Jiebo Luo.
      International Joint Conference on Artificial Intelligence (IJCAI), 2020. (AC Rate = 12.6%)
    25. Learning Task-aware Local Representations for Few-shot Learning
      Chuanqi Dong, Wenbin Li, Jing Huo, Zheng Gu, Yang Gao.
      International Joint Conference on Artificial Intelligence (IJCAI), 2020. (AC Rate = 12.6%)
    26. Joint Multi-view 2D Convolutional Neural Networks for 3D Object Classification
      Jinglin Xu, Xiangsen Zhang, Wenbin Li, Xinwang Liu, Junwei Han.
      International Joint Conference on Artificial Intelligence (IJCAI), 2020. (AC Rate = 12.6%)
    27. Biased Feature Learning for Occlusion Invariant Face Recognition
      Changbin Shao, Jing Huo, Lei Qi, Zhen-Hua Feng, Wenbin Li, Chuanqi Dong, Yang Gao.
      International Joint Conference on Artificial Intelligence (IJCAI), 2020. (AC Rate = 12.6%)
    28. Deep Embedded Complementary and Interactive Information for Multi-view Classification
      Jinglin Xu, Wenbin Li, Xinwang Liu, Dingwen Zhang, Ji Liu, Junwei Han.
      AAAI Conference on Artificial Intelligence (AAAI), 2020. (AC Rate = 20.6%)
    29. Layerwise Sparse Coding for Pruned Deep Neural Networks with Extreme Compression Ratio
      Xiao Liu, Wenbin Li, Jing Huo, Lili Yao, Yang Gao.
      AAAI Conference on Artificial Intelligence (AAAI), 2020. (AC Rate = 20.6%)
    30. Revisiting Local Descriptor based Image-to-Class Measure for Few-shot Learning
      Wenbin Li, Lei Wang, Jinglin Xu, Jing Huo, Yang Gao, Jiebo Luo.
      IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2019. (AC Rate = 25.2%)
    31. Distribution Consistency based Covariance Metric Networks for Few-shot Learning
      Wenbin Li, Jinglin Xu, Jing Huo, Lei Wang, Yang Gao, Jiebo Luo.
      AAAI Conference on Artificial Intelligence (AAAI), 2019. (AC Rate = 16.2%) [Oral]

    Selected Preprints

    1. A Benchmark for Ultra-High-Resolution Remote Sensing MLLMs
      Yunkai Dang#, Meiyi Zhu#, Donghao Wang, Yizhuo Zhang, Jiacheng Yang, Qi Fan, Yuekun Yang, Wenbin Li*,
      Feng Miao, Yang Gao.
      arXiv:2512.17319, 2025
    2. Robust Dataset Distillation by Matching Adversarial Trajectories
      Wei Lai, Tianyu Ding, ren dongdong, Lei Wang, Jing Huo, Yang Gao, Wenbin Li*.
      arXiv:2503.12069, 2025
    3. A Unified Framework for Contrastive Learning from a Perspective of Affinity Matrix
      Wenbin Li, Meihao Kong, Xuesong Yang, Lei Wang, Jing Huo, Yang Gao and Jiebo Luo.
      arXiv:2211.14516, 2022

    Datasets & Frameworks

    1. 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.
    2. 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.
    3. 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.

    Student Mentoring & Supervision

    (†, ‡, †† indicate co-supervision with Prof. Yang Gao, Yuekun Yang, and Cong Wang, respectively)
    Name Department Starting Graduated Program First Job
    Dongdong Ren† ESE 2021.9 2025.6 Ph.D. 吉利汽车
    Yuhang Zhu† ESE 2023.9 in progress Ph.D. -
    Zhiping Wu† ESE 2024.9 in progress Ph.D. -
    Yunchen Wu† CS 2024.9 in progress Ph.D. -
    Borui Kang IS 2025.9 in progress Ph.D. -
    Liuyuan Wen IS 2025.9 in progress Ph.D. -
    Yunkai Dang IS 2025.9 in progress Ph.D. -
    Jiacheng Yang IS 2025.9 in progress Ph.D. -
    Jizhe Dou‡ IS 2025.9 in progress Ph.D. -
    Tianjiao Jiao‡ IS 2025.9 in progress Ph.D. -
    Feifan Tian†† IS 2025.9 in progress Ph.D. -
    Ziyi Wang CS 2020.9 2023.6 M.S. 壁仞科技
    Xuesong Yang CS 2020.9 2023.6 M.S. 壁仞科技
    Boyao Shi CS 2020.9 2023.6 M.S. 农业银行
    Jiayan Chen CS 2020.9 2023.6 M.S. 华为
    Mengying Sun CS 2021.9 2024.6 M.S. 美团
    Yakun Xiong CS 2021.9 2024.6 M.S. 地平线
    Zhichen Fan CS 2021.9 2024.6 M.S. 中国电信
    Wenxiao Deng CS 2022.9 2025.6 M.S. 阿里淘天
    Yiyang Chen CS 2023.9 in progress M.S. -
    Wei Lai CS 2023.9 in progress M.S. -
    Bing Li CS 2023.9 in progress M.S. -
    Shangge Liu CS 2024.9 in progress M.S. -
    Jiaxin Li IS 2025.9 in progress M.S. -
    Zhengjie Zhang IS 2025.9 in progress M.S. -
    Yang Chen IS 2025.9 in progress M.S. -

    Teaching

    Awards

    Services

    © 2025 Wenbin Li | Last update: Oct. 2025