Bio

I am an Assistant Professor (Principal investigator, PhD supervisor) in the Department of Statistics and Data Science at SUSTech (China, Shenzhen). I obtained my Ph.D. degree at the School of Computer Science and Engineering, Nanyang Technological University, supervised by Prof. Bo An. During my Ph.D, I was fortunate to work as a visiting scholar in the group of Prof. Sharon Yixuan Li at the University of Wisconsin Madison in 2022. Previously I spent a wonderful year as a research assistant in the Institute for Interdisciplinary Information Sciences at Tsinghua University. Prior to that, I received my B.E. in Software Engineering from Huazhong University of Science and Technology in 2016.

My research interest falls in the scope of reliable machine learning (uncertainty estimation), and its applications in data optimization and privacy. Generally, we expect deep learning models to produce precise estimation of their uncertainty in predictions, using the form of probabilities (confidences in ML) or conformal prediction sets. Besides, my research is closely related to data-centric machine learning and foundation model, like data quality and efficiency. We are also interested in the statistic theory of data selection in machine learning, which provides theoretical principles to guide the data optimization in machine learning workflows.

We are always actively looking for Postdocs, PhDs, and RA/interns to join our research.

News

February 2025
Three papers are accepted by ICML 2025. Congratulations to Shuoyuan and Hao Zeng!
February 2025
I will be serving as a Area Chair for NeurIPS 2025.
May 2024
Four papers are accepted by ICLR 2025. Congratulations to Kangdao, Wenyu, and Hengxiang!
January 2025
I will be serving as a SPC for IJCAI 2025.
November 2024
I will be serving as a Area Chair for ICML 2025.
May 2024
Two papers are accepted by NeurIPS 2024. Congratulations to Hongfu Gao!
July 2024
We release a Chinese Content Moderation Benchmark for LLMs: ChineseSafe-Benchmark.
May 2024
I am accepting Mphil and PhD applications (2025 fall). I am always looking for highly-motivated research interns, RAs and PostDocs to join our research (refer to this page).
May 2024
Four papers are accepted by ICML 2024.
March 2024
I will be serving as a Area Chair for NeurIPS 2024.
January 2024
Three papers are accepted by ICLR 2024 (Two are Spotlights).
December 2023
We release a Python toolbox for conformal prediction research TorchCP.
October 2022
I am honored to be recognized as Top Reviewers in NeurIPS 2022.
May 2022
Two papers are accepted by ICML 2022 (Accept rate: 21.9%).
October 2021
I am honored to receive NeurIPS 2021 Outstanding Reviewer Award (top 8% of reviewers).

Working papers

( * Corresponding author )
Your Pre-trained LLM is Secretly an Unsupervised Confidence Calibrator
Beier Luo, Shuoyuan Wang, Yixuan Li, Hongxin Wei *
Semi-Supervised Conformal Prediction With Unlabeled Nonconformity Score
Xuanning Zhou, Hao Zeng, Xiaobo Xia, Bingyi Jing, Hongxin Wei *
Efficient Membership Inference Attacks by Bayesian Neural Network
Zhenlong Liu, Wenyu Jiang, Feng Zhou, Hongxin Wei *
Exploring Imbalanced Annotations for Effective In-Context Learning
Hongfu Gao, Feipeng Zhang, Hao Zeng, Deyu Meng, Bingyi Jing, Hongxin Wei *
Robust Online Conformal Prediction under Uniform Label Noise
Huajun Xi, Kangdao Liu, Hao Zeng, Wenguang Sun, Hongxin Wei *
Defending Membership Inference Attacks via Privacy-aware Sparsity Tuning
Qiang Hu, Hengxiang Zhang, Hongxin Wei *
TorchCP: A Library for Conformal Prediction based on PyTorch
Hongxin Wei, Jianguo Huang

Selected Publications

( * Corresponding author; Equal contribution)
Does Confidence Calibration Help Conformal Prediction?
TMLR
Huajun Xi, Jianguo Huang, Kangdao Liu, Lei Feng, Hongxin Wei *
Understanding and Mitigating Miscalibration in Prompt Tuning for Vision-Language Models
ICML 2025
Shuoyuan Wang, Yixuan Li, Hongxin Wei *
Parametric Scaling Law of Tuning Bias in Conformal Prediction
ICML 2025
Hao Zeng, Kangdao Liu, Bingyi Jing, Hongxin Wei *
How Contaminated Is Your Benchmark? Quantifying Dataset Leakage in Large Language Models with Kernel Divergence
ICML 2025
Hyeong Kyu Choi, Maxim Khanov, Hongxin Wei, Yixuan Li
Exploring Learning Complexity for Downstream Data Pruning
ICLR 2025
Wenyu Jiang, Zhenlong Liu, Zejian Xie, Songxin Zhang, Bingyi Jing, Hongxin Wei *
Fine-tuning can Help Detect Pretraining Data from Large Language Models
ICLR 2025
Hengxiang Zhang, Songxin Zhang, Bingyi Jing, Hongxin Wei *
On the Noise Robustness of In-Context Learning for Text Generation
NeurIPS 2024
Hongfu Gao, Feipeng Zhang, Wenyu Jiang, Jun Shu, Feng Zheng, Hongxin Wei*
Open-Vocabulary Calibration for Vision-Language Models
ICML 2024
Shuoyuan Wang, Jindong Wang, Guoqing Wang, Bob Zhang, Kaiyang Zhou, Hongxin Wei *
Conformal Prediction for Deep Classifier via Label Ranking
ICML 2024
Jianguo Huang, Huajun Xi, Linjun Zhang, Huaxiu Yao, Yue Qiu, Hongxin Wei *
Mitigating Privacy Risk in Membership Inference by Convex-Concave Loss
ICML 2024
Zhenlong Liu, Lei Feng, Huiping Zhuang, Xiaofeng Cao, Hongxin Wei *
DOS: Diverse Outlier Sampling for Out-of-Distribution Detection
ICLR 2024
Wenyu Jiang, Hao Cheng, MingCai Chen, Chongjun Wang, Hongxin Wei *
Consistent Multi-Class Classification from Multiple Unlabeled Datasets
ICLR 2024 (Spotlight)
Zixi Wei, Senlin Shu, Yuzhou Cao, Hongxin Wei, Bo An, Lei Feng
Optimization-Free Test-Time Adaptation for Cross-Person Activity Recognition.
IMWUT/Ubicomp 2024
Shuoyuan Wang, Jindong Wang, HuaJun Xi, Bob Zhang, Lei Zhang, Hongxin Wei
On the Importance of Feature Separability in Predicting Out-Of-Distribution Error.
NeurIPS 2023
Renchunzi Xie, Hongxin Wei *, Lei Feng, Yuzhou Cao, Bo An
Mitigating Memorization of Noisy Labels by Clipping the Model Prediction
ICML 2023
Hongxin Wei, Huiping Zhuang, Renchunzi Xie, Lei Feng, Gang Niu, Bo An, Yixuan Li
Mitigating Neural Network Overconfidence with Logit Normalization
ICML 2022
Hongxin Wei, Renchunzi Xie, Hao Cheng, Lei Feng, Bo An, Yixuan Li
Open-Sampling: Exploring Out-of-Distribution data for Re-balancing Long-tailed datasets
ICML 2022
Hongxin Wei, Lue Tao, Renchunzi Xie, Lei Feng, Bo An
Deep Learning from Multiple Noisy Annotators as A Union
TNNLS
Hongxin Wei, Renchunzi Xie, Lei Feng, Bo An
GearNet: Stepwise Dual Learning for Weakly Supervised Domain Adaptation
AAAI 2022
Renchunzi Xie, Hongxin Wei *, Lei Feng, Bo An
Open-set Label Noise Can Improve Robustness Against Inherent Label Noise
NeurIPS 2021
Hongxin Wei, Lue Tao, Renchunzi Xie, Bo An
Commission Fee is not Enough: A Hierarchical Reinforced Framework for Portfolio Management
AAAI 2021
Rundong Wang , Hongxin Wei , Bo An, Zhouyan Feng, Jun Yao
Combating noisy labels by agreement: A joint training method with co-regularization
CVPR 2020
Hongxin Wei, Lei Feng, Xiangyu Chen, Bo An

Research Group

Hao Zeng
Postdoc (co-supervised with Prof. B. Jing)
PhD degree from Xiamen University
Shuoyuan Wang
Ph.D. student
Master degree from University of Macau
Hengxiang Zhang
Ph.D. student
Master degree from UESTC
Zhenlong Liu
Ph.D. student
Bachelor degree from SUSTech
Beier Luo
Master student
Bachelor degree from SUSTech
Cong Ding
Master student
Now Intern at Tencent
Haotian Zhang
Research Intern,
Master student at DUT
Huajun Xi
Research Assistant,
Bachelor degree from SUSTech
Wenbo Liao
Research Intern,
PhD student at CUHK
Qingyang Hao
Research Assistant,
Master degree from THU(SZ)
Xuanning Zhou
Research Intern,
Undergraduate student at HITSZ
Yuhan Li
Research Intern,
Undergraduate student at SDU(Weihai)

Past Members

Jianguo Huang
Research Intern (2023-2024)
Now PhD student at NTU, Singapore
Qiqi Tao
Research Intern (2024),
Now PhD student at SUTD, Singapore
Hongfu Gao
Research Intern (2024),
Now PhD student at NUS
Jianqing Song
Research Intern (2024),
PhD student at Nanjing University
Wenyu Jiang
Research Intern (2023-2024)
Ph.D. student at Nanjing University.
Kangdao Liu
Research Intern (2024)
Ph.D. student at University of Macau
Qiang Hu
Now PhD student at SUNY Buffalo
Undergraduate student at SUSTech