Research

The long-term research goal is to build robust models for modern AI, such as pre-trained models and large models. We create new theory, algorithms, applications, and open-sourced library to achieve our goal. These days, we are specifically interested in robustness in large language models (LLMs).

Our research consists of the following topics with selected publications: [View by year]

Out-of-distribution (Domain) generalization and adaptation for distribution shift

• [ICLR’23] Out-of-distribution Representation Learning for Time Series Classification. Wang Lu, Jindong Wang, Xinwei Sun, Yiqiang Chen, and Xing Xie.
• [KDD’23] Domain-Specific Risk Minimization for Out-of-Distribution Generalization. YiFan Zhang, Jindong Wang, Jian Liang, Zhang Zhang, Baosheng Yu, Liang Wang, Xing Xie, and Dacheng Tao.
• [KDD’23] Generalizable Low-Resource Activity Recognition with Diverse and Discriminative Representation Learning. Xin Qin, Jindong Wang, Shuo Ma, Wang Lu, Yongchun Zhu, Xing Xie, Yiqiang Chen.
• [ACL’23 findings] GLUE-X: Evaluating Natural Language Understanding Models from an Out-of-distribution Generalization Perspective. Linyi Yang, Shuibai Zhang, Libo Qin, Yafu Li, Yidong Wang, Hanmeng Liu, Jindong Wang, Xing Xie, Yue Zhang.
• [TKDE’22] Generalizing to Unseen Domains: A Survey on Domain Generalization. Jindong Wang, Cuiling Lan, Chang Liu, Yidong Ouyang, Tao Qin, Wang Lu, Yiqiang Chen, Wenjun Zeng, and Philip S. Yu.
• [TMLR’22] Domain-invariant Feature Exploration for Domain Generalization. Wang Lu, Jindong Wang, Haoliang Li, Yiqiang Chen, and Xing Xie.
• [UbiComp’22] Semantic-Discriminative Mixup for Generalizable Sensor-based Cross-domain Activity Recognition. Wang Lu, Jindong Wang, Yiqiang Chen, Sinno Pan, Chunyu Hu, and Xin Qin.
• [NeurIPS’21] Learning causal semantic representation for out-of-distribution prediction. Chang Liu, Xinwei Sun, Jindong Wang , Haoyue Tang, Tao Li, Tao Qin, Wei Chen, and Tie-Yan Liu.
• [CIKM’21] Adarnn: Adaptive learning and forecasting of time series. Yuntao Du, Jindong Wang, Wenjie Feng, Sinno Pan, Tao Qin, Renjun Xu, and Chongjun Wang.
• [TNNLS’20, 300 + citations] Deep subdomain adaptation network for image classification. Yongchun Zhu, Fuzhen Zhuang, Jindong Wang, Guolin Ke, Jingwu Chen, Jiang Bian, Hui Xiong, and Qing He.
• [ACMMM’18, 400+ citations] Visual domain adaptation with manifold embedded distribution alignment. Jindong Wang, Wenjie Feng, Yiqiang Chen, Han Yu, Meiyu Huang, and Philip S Yu.
• [ICDM’17, 400+ citations] Balanced distribution adaptation for transfer learning. Jindong Wang, Yiqiang Chen, Shuji Hao, Wenjie Feng, and Zhiqi Shen.
• Open-source:
  • Transfer learning
  • robustlearn: A unified repo for robust machine learning, such as OOD and adversarial robustness: robustlearn
  • PandaLM: PandaLM

Semi-supervised learning for low-resource learning

• [ICLR’23] FreeMatch: Self-adaptive Thresholding for Semi-supervised Learning. Yidong Wang, Hao Chen, Qiang Heng, Wenxin Hou, Yue Fan, Zhen Wu, Jindong Wang, Marios Savvides, Takahiro Shinozaki, Bhiksha Raj, Bernt Schiele, and Xing Xie.
• [ICLR’23] SoftMatch: Addressing the Quantity-Quality Tradeoff in Semi-supervised Learning. Hao Chen, Ran Tao, Yue Fan, Yidong Wang, Jindong Wang, Bernt Schiele, Xing Xie, Bhiksha Raj, and Marios Savvides.
• [NeurIPS’22] USB: A Unified Semi-supervised Learning Benchmark. Yidong Wang, Hao Chen, Yue Fan, Wang Sun, Ran Tao, Wenxin Hou, Renjie Wang, Linyi Yang, Zhi Zhou, Lan-Zhe Guo, Heli Qi, Zhen Wu, Yu-Feng Li, Satoshi Nakamura, Wei Ye, Marios Savvides, Bhiksha Raj, Takahiro Shinozaki, Bernt Schiele, Jindong Wang, Xing Xie, and Yue Zhang.
• [TASLP’22] Exploiting Adapters for Cross-lingual Low-resource Speech Recognition. Wenxin Hou, Han Zhu, Yidong Wang, Jindong Wang, Tao Qin, Renjun Xu, and Takahiro Shinozaki.
• [NeurIPS’21, 200+ citations] Flexmatch: Boosting semi-supervised learning with curriculum pseudo labeling. Bowen Zhang, Yidong Wang, Wenxin Hou, Hao Wu, Jindong Wang, Manabu Okumura, and Takahiro Shinozaki.
• Open-source:
  • USB: A unified semi-supervised learning toolbox for CV, NLP, and Audio: USB
  • A unified Pytorch-based semi-supervised learning library: **TorchSSL

Safe transfer learning for security

• [arXiv’23] On the Robustness of ChatGPT: An Adversarial and Out-of-distribution Perspective. Jindong Wang, Xixu Hu, Wenxin Hou, Hao Chen, Runkai Zheng, Yidong Wang, Linyi Yang, Haojun Huang, Wei Ye, Xiubo Geng, Binxin Jiao, Yue Zhang, and Xing Xie.
• [ICSE’22] ReMoS: Reducing Defect Inheritance in Transfer Learning via Relevant Model Slicing. Ziqi Zhang, Yuanchun Li, Jindong Wang, Bingyan Liu, Ding Li, Xiangqun Chen, Yao Guo, and Yunxin Liu.
• [IEEE TBD’22] Personalized Federated Learning with Adaptive Batchnorm for Healthcare. Wang Lu, Jindong Wang, Yiqiang Chen, Xin Qin, Renjun Xu, Dimitrios Dimitriadis, and Tao Qin.
• [TKDE’22] Unsupervised deep anomaly detection for multi-sensor time-series signals. Yuxin Zhang, Yiqiang Chen, Jindong Wang, and Zhiwen Pan.
• [IntSys’22, 400+ citations] Fedhealth: A federated transfer learning framework for wearable healthcare. Yiqiang Chen, Xin Qin, Jindong Wang, Chaohui Yu, and Wen Gao.
• Open-source:
  • PersonalizedFL: a personalized federated learning libraty: PersonalizedFL
  • robustlearn: A unified repo for robust machine learning, such as OOD and adversarial robustness: robustlearn

Imbalanced learning for long-tailed tasks

• [arXiv’23] Exploring Vision-Language Models for Imbalanced Learning. Yidong Wang, Zhuohao Yu, Jindong Wang, Qiang Heng, Hao Chen, Wei Ye, Rui Xie, Xing Xie, Shikun Zhang.
• [ACML’22] Margin Calibration for Long-Tailed Visual Recognition. Yidong Wang, Bowen Zhang, Wenxin Hou, Zhen Wu, Jindong Wang, and Takahiro Shinozaki.
• Open-source:
  • Imbalance-VLM: a library for imbalanced learning in vision-language models. [Imbalance-VLM]

Miscellaneous

1. An easy-to-use speech recognition toolkit based on Espnet: EasyESPNet
2. Leading the transfer learning tutorial (迁移学习简明手册) on Github: Tutorial
3. I’m also leading other popular research projects: Machine learning, Activity recognition
4. I started a software studio Pivot Studio and made many applications in 2010-2014: Our applications