Jianing Zhu

Postdoctoral Fellow @ UT Austin

VITA Group Chandra Department of ECE
Cockrell School of Engineering,

also affiliated with Good Systems Challenge,
Ethical AI at UT Austin,

University of Texas at Austin,
Austin, Texas, USA

[Google Scholar] [Github] [Twitter] [LinkedIn] [CV]

E-mail: jianing.zhu@austin.utexas.edu or zhujianing9810@gmail.com

Address: 3rd floor at EER, UT Austin, 2501 Speedway, Austin, TX 78712, USA

About Me

I am a Postdoctoral Fellow at VITA group of UT Austin, working with Prof. Atlas Wang. I received my PhD at Trustworthy Machine Learning and Reasoning (TMLR) Group in the Department of Computer Science, Hong Kong Baptist University, advised by Dr. Bo Han. I was a visiting student researcher at CMU MLD, fortunately working with Prof. Pradeep Ravikumar, and I was a research intern at RIKEN AIP, fortunately working with Dr. Gang Niu and Prof. Masashi Sugiyama. Before that, I received my B.Eng. degree in Computer Science and Technology (Top-notch Student Program) from Sichuan University in 2021.

My research agenda is AI Forgetting: Intentional and Unintentional. I study how AI systems stay reliable over their operational lifetime as they learn, adapt, and accumulate memory, and how they should manage what they retain, erase, and expose. The agenda runs along three threads:

Building Robustness Against Degradation. How can models and agents withstand adversarial stress by design?
[AgentHijack] [SFAT] [GAIRAT] [ATCL]
Diagnosing Failure and Degradation. Where, when, and why do evolving agents drift, forget, and destabilize?
[AgentForesight] [CoVer] [NoRa] [Unleashing Mask]
Governing What Models Retain. What should a system retain, erase, and expose, and can an auditor verify that it did?
[AgentTriMem] [TARF] [CRU] [Copyright Detective] [DeepInception]

These threads serve one goal: the lifecycle reliability of evolving AI systems. Reliability is no longer a property to be verified once at deployment; it erodes as a system learns, adapts, and accumulates memory. My goal is to make resilience a first-class property of evolving AI systems, so that they can be responsibly and safely deployed for societal benefit in high-stakes, real-world applications. At UT Austin, I work with several junior researchers and students on agent reliability and evaluation. If this agenda interests you, feel free to get in touch.

Remarkable Projects

(Click the images for more details)

Research Experience

Postdoctoral Fellow, 2025.09 - Present
VITA Group, Chandra Department of Electrical and Computer Engineering,
Cockrell School of Engineering, University of Texas at Austin
Advised by Prof. Atlas Wang
Austin, Texas, United States
Visiting PhD Student, 2025.01 - 2025.06
Statistical & Symbolic Learning (Neuro-Symbolic AI) Group, Machine Learning Department,
School of Computer Science, Carnegie Mellon University
Advised by Prof. Pradeep Ravikumar
Pittsburgh, Pennsylvania, United States
Research Intern, 2023.12 - 2024.05
Imperfect Information Learning Team, RIKEN AIP
Advised by Dr. Gang Niu and Prof. Masashi Sugiyama
RIKEN, Tokyo, Japan

Education

Ph.D. student, 2021.09 - 2025.07
TMLR Group, Department of Computer Science, Faculty of Science
Hong Kong Baptist University (HKBU), Hong Kong SAR
B.Eng., 2017.09 - 2021.06
College of Computer Science (Top-notched Student Program)
Sichuan University (SCU), Chengdu, China

Selected Publication

(* indicates equal contribution, full list can refer to Google Scholar)

Conference Paper

J. Sun, J. Zhu, Y. Li, T. Liu, X. Hu, B. Han
AgentHijack: Benchmarking Computer Use Agent Robustness to Common Environment Corruptions.
In International Conference on Machine Learning (ICML 2026), Brazil, 2026.
[PDF] [BibTeX]

@inproceedings{sun2026agenthijack,
  title     = {AgentHijack: Benchmarking Computer Use Agent Robustness to Common Environment Corruptions},
  author    = {Jingwei Sun and Jianing Zhu and Yuanyi Li and Tongliang Liu and Xia Hu and Bo Han},
  booktitle = {Forty-third International Conference on Machine Learning},
  year      = {2026},
  url       = {https://arxiv.org/abs/2605.25707}
}

Z. Zhang*, J. Zhu*, X. Ge*, Z. Zhao*, Z. Zhou, X. Xuan, X. Feng, J. Yao, B. Han.
Co-rewarding: Stable Self-supervised RL for Eliciting Reasoning in Large Language Models.
In International Conference on Learning Representations (ICLR 2026), Brazil, 2026.
[PDF] [BibTeX]

@inproceedings{zhang2026co,
        title={Co-Reward: Self-supervised Reinforcement Learning for Large Language Model Reasoning via Contrastive Agreement},
        author={Zhang*, Zizhuo and Jianing Zhu* and Ge*, Xinmu and Zhao*, Zihua and Zhou, Zhanke and Li, Xuan and Feng, Xiao and Yao, Jiangchao and Han, Bo},
        booktitle={The Fourteenth International Conference on Learning Representations},
        year={2026}
}

J. Zhu, B. Han, J. Yao, J. Xu, G. Niu, M. Sugiyama.
Decoupling the Class Label and the Target Concept in Machine Unlearning.
In International Conference on Learning Representations (ICLR 2026), Brazil, 2026.
[PDF] [BibTeX]

@inproceedings{zhu2026decoupling,
        title={Decoupling the Class Label and the Target Concept in Machine Unlearning}, 
        author={Jianing Zhu and Bo Han and Jiangchao Yao and Jianliang Xu and Gang Niu and Masashi Sugiyama},
        booktitle={The Fourteenth International Conference on Learning Representations},
        year={2026}
}

J. Sun*, J. Zhu*, J. Yao, G. Niu, M. Sugiyama, B. Han
Bilateral Information-aware Test-time Adaptation for Vision-Language Models.
In International Conference on Learning Representations (ICLR 2026), Brazil, 2026.
[PDF] [BibTeX]

@inproceedings{sun2026bilateral,
        title={Bilateral Information-aware Test-time Adaptation for Vision-Language Models},
        author={Sun*, Jie and Zhu*, Jianing and Yao, Jiangchao and Niu, Gang and Sugiyama, Masashi and Han, Bo},
        booktitle={The Fourteenth International Conference on Learning Representations},
        year={2026}
}

Z. Zhang, Q. Wang, S. Ye, J. Zhu, J. Yao, B. Han, M. Sugiyama
Towards Understanding Valuable Preference Data for Large Language Model Alignment.
In International Conference on Learning Representations (ICLR 2026), Brazil, 2026.
[PDF] [BibTeX]

@inproceedings{zhang2026towards,
        title={Towards Understanding Valuable Preference Data for Large Language Model Alignment},
        author={Zhang, Zizhuo and Wang, Qizhou and Ye, Sheng and Zhu, Jianing and Yao, Jiangchao and Han, Bo and Sugiyama, Masashi},
        booktitle={The Fourteenth International Conference on Learning Representations},
        year={2026}
}

Q. Wang, B. Han, Y. Pu, J. Zhu, T. Liu, M. Sugiyama.
Towards Effective Evaluations and Comparisons for LLM Unlearning Methods
In International Conference on Learning Representations (ICLR 2025), Singapore, 2025.
[PDF] [BibTeX]

@misc{wang2024unlearning,
        title={Unlearning with Control: Assessing Real-world Utility for Large Language Model Unlearning}, 
        author={Qizhou Wang and Bo Han and Puning Yang and Jianing Zhu and Tongliang Liu and Masashi Sugiyama},
        year={2024},
        eprint={2406.09179},
        archivePrefix={arXiv},
}

B. Zhang*, J. Zhu*, Z. Wang, T. Liu, B. Du, B. Han.
What If the Input is Expanded in OOD Detection?
In Advances in Neural Information Processing Systems (NeurIPS 2024), Vancouver, 2024.
[PDF] [Code] [BibTeX]

@inproceedings{zhang2024what,
        title={What If the Input is Expanded in OOD Detection?},
        author={Zhang, Boxuan and Zhu, Jianing and Wang, Zengmao and Liu, Tongliang and Du, Bo and Han, Bo},
        booktitle={The Thirty-Eighth Annual Conference on Neural Information Processing Systems},
        year={2024},
}

Y. Geng, J. Zhu, J. Yao, B. Han.
Self-Calibrated Tuning of Vision-Language Models for Out-of-Distribution Detection
In Advances in Neural Information Processing Systems (NeurIPS 2024), Vancouver, 2024.
[PDF] [Code] [BibTeX]

@inproceedings{geng2024self,
        title={Self-Calibrated Tuning of Vision-Language Models for Out-of-Distribution Detection},
        author={Geng, Yu and Zhu, Jianing and Yao, Jiangchao and Han, Bo},
        booktitle={The Thirty-Eighth Annual Conference on Neural Information Processing Systems},
        year={2024},
}

Z. Zhou, R. Tao, J. Zhu, Y. Luo, Z. Wang, B. Han.
Benchmarking the Reasoning Robustness against Noisy Rationales in Chain-of-thought Prompting
In Advances in Neural Information Processing Systems (NeurIPS 2024), Vancouver, 2024.
[PDF] [Code] [BibTeX]

@inproceedings{zhou2024benchmarking,
        title={Benchmarking the Reasoning Robustness against Noisy Rationales in Chain-of-thought Prompting},
        author={Zhou, Zhanke and Tao, Rong and Zhu, Jianing and Luo, Yiwen and Wang, Zengmao and Han, Bo},
        booktitle={The Thirty-Eighth Annual Conference on Neural Information Processing Systems},
        year={2024},
}

J. Zhu, G. Yu, J. Yao, T. Liu, G. Niu, M. Sugiyama, B. Han.
Diversified Outlier Exposure for Out-of-Distribution Detection via Informative Extrapolation.
In Advances in Neural Information Processing Systems (NeurIPS 2023), New Orleans, 2023.
[PDF] [Code] [BibTeX]

@inproceedings{
        zhu2023diversified,
        title={Diversified Outlier Exposure for Out-of-Distribution Detection via Informative Extrapolation},
        author={Jianing Zhu, Geng Yu, Jiangchao Yao, Tongliang Liu, Gang Niu, Masashi Sugiyama, Bo Han},
        booktitle={Thirty-seventh Conference on Neural Information Processing Systems},
        year={2023}
}

J. Zhu, H. Li, J. Yao, T. Liu, J. Xu, B Han.
Unleashing Mask: Explore the Intrinsic Out-of-distribution Detection Capability.
In International Conference on Machine Learning (ICML 2023), Hawaii, 2023.
[PDF] [Code] [BibTeX]

@inproceedings{zhu2023unleashing,
        title={Unleashing Mask: Explore the Intrinsic Out-of-distribution Detection Capability}, 
        author={Jianing Zhu and Hengzhuang Li and Jiangchao Yao and Tongliang Liu and Jianliang Xu and Bo Han},
        booktitle = {International Conference on Machine Learning},
        year = {2023}
}

J. Zhu, J. Yao, T. Liu, Q. Yao, J. Xu, and B. Han.
Combating Exacerbated Heterogeneity for Robust Models in Federated Learning.
In International Conference on Learning Representations (ICLR 2023), Kigali, 2023.
[PDF] [Code] [BibTeX]

@inproceedings{
    zhu2023combating,
    title={Combating Exacerbated Heterogeneity for Robust Models in Federated Learning},
    author={Jianing Zhu and Jiangchao Yao and Tongliang Liu and Quanming Yao and Jianliang Xu and Bo Han},
    booktitle={The Eleventh International Conference on Learning Representations},
    year={2023},
    url={https://openreview.net/forum?id=eKllxpLOOm}
}

J. Zhou*, J. Zhu*, J. Zhang, T. Liu, G. Niu, B. Han, M. Sugiyama.
Adversarial Training with Complementary Labels: On the Benefit of Gradually Informative Attacks.
In Advances in Neural Information Processing Systems (NeurIPS 2022), Online, 2022.
[PDF] [Code] [BibTeX] [Spotlight]

@inproceedings{
    zhou2022adversarial,
    title={Adversarial Training with Complementary Labels: On the Benefit of Gradually Informative Attacks},
    author={Jianan Zhou and Jianing Zhu and Jingfeng Zhang and Tongliang Liu and Gang Niu and Bo Han and Masashi Sugiyama},
    booktitle={Advances in Neural Information Processing Systems},
    editor={Alice H. Oh and Alekh Agarwal and Danielle Belgrave and Kyunghyun Cho},
    year={2022},
    url={https://openreview.net/forum?id=s7SukMH7ie9}
}

J. Zhu, J. Yao, B. Han, J. Zhang, T. Liu, G. Niu, J. Zhou, J. Xu, H. Yang.
Reliable Adversarial Distillation with Unreliable Teachers.
In International Conference on Learning Representations (ICLR 2022), Online, 2022.
[PDF] [Code] [BibTeX]

@inproceedings{
    zhu2022reliable,
    title={Reliable Adversarial Distillation with Unreliable Teachers},
    author={Jianing Zhu and Jiangchao Yao and Bo Han and Jingfeng Zhang and Tongliang Liu and Gang Niu and Jingren Zhou and Jianliang Xu and Hongxia Yang},
    booktitle={International Conference on Learning Representations},
    year={2022},
    url={https://openreview.net/forum?id=u6TRGdzhfip}
}

J. Zhang, J. Zhu, G. Niu, B. Han, M. Sugiyama, and M. Kankanhalli.
Geometry-aware Instance-reweighted Adversarial Training.
In International Conference on Learning Representations (ICLR 2021), Online, 2021.
[PDF] [Code] [BibTeX] [Oral]

@inproceedings{
    zhang2021geometryaware,
    title={Geometry-aware Instance-reweighted Adversarial Training},
    author={Jingfeng Zhang and Jianing Zhu and Gang Niu and Bo Han and Masashi Sugiyama and Mohan Kankanhalli},
    booktitle={International Conference on Learning Representations},
    year={2021},
    url={https://openreview.net/forum?id=iAX0l6Cz8ub}
}

Journal Paper

J. Zhu, B. Han, J. Yao, T. Liu, Q. Yao, and J. Xu.
POPS: Recovering Unlearned Multi-Modality Knowledge in MLLMs with Prompt-Optimized Parameter Shaking.
Transactions on Machine Learning Research (TMLR), 2026.
[PDF] [Code] [BibTeX]

@article{li2026pops,
  title={POPS: Recovering Unlearned Multi-Modality Knowledge in MLLMs with Prompt-Optimized Parameter Shaking},
  author={Li, Zhangheng and Zhu, Jianing and Hong, Junyuan and Eum, Sungmin and Hu, Shuowen and You, Suya and Wang, Zhangyang},
  journal={Transactions on Machine Learning Research},
  year={2026}
}

J. Zhu, B. Han, J. Yao, T. Liu, Q. Yao, and J. Xu.
Slack Federated Adversarial Training.
IEEE Transactions on Pattern Analysis and Machine Intelligence (TPAMI), 2025.
[PDF] [Code] [BibTeX]

@inproceedings{
    zhu2025slack,
    title={Slack Federated Adversarial Training},
    author={Jianing Zhu and Bo Han and Jiangchao Yao and Tongliang Liu and Quanming Yao and Jianliang Xu},
    booktitle={IEEE Transactions on Pattern Analysis and Machine Intelligence},
    year={2025},
}

Honors & Award

Anthropic External Researcher Access Program (2026)
TrustAI Rising Star Award (2025)
NeurIPS Top Reviewer, NeurIPS (2023, 2025)
ICML Best Reviewer, ICML (2021,2024)
ICLR Notable Reviewer, ICLR (2025)
Yakun Scholarship Scheme for Mainland Postgraduate Students, HKBU (2024)
Computer Science Department Research Excellence Award, HKBU (2023)
Computer Science Department RPg Performance Award, HKBU (2022, 2023, 2024)
Excellent Teaching Assistant Performance Award, HKBU (2022, 2023)
Nominated for Hong Kong PhD Fellowship Scheme (HKPFS), HKBU (2021)
National Scholarship, Ministry of Education (2018)
University Scholarship, SCU (2018, 2019, 2020)

Academic Talk

Surgical Machine Unlearning: Decoupling Targets, Diagnosing Fragility, and Latent Intervention @ UIUC, Online, Mar, 2026.
Unlearning Sensitive Information @ Texas Symposium, Mar, 2026.
Surgical Machine Unlearning: Decoupling Targets, Diagnosing Fragility, and Latent Intervention @ NTU, Singapore, Jan, 2026.
Tutorial of “Handling Out-of-Distribution Data in the Open World” @ AAAI 2026, Singapore, Jan, 2026.
Lecture of “Handling OOD Data” @ SJTU Course of Deep Learning for Advanced Computer Vision, Online, Nov, 2025.
How AI Leaks Information — and How It Can Forget @ UT Austin Digital Trust Symposium, Nov. 2025.
Invited Talk about Trustworthy Machine Learning @ SCU Top-notched Program, Online, Dec.09, 2023.
Youth PhD Talk for NeurIPS 2023 @ AI Time, Online, Nov.02, 2023. [Video]
Invited Talk about SFAT @ XMU ASC, Online, Aug.14, 2023.
Youth PhD Talk for ICLR 2023 @ AI Time, Online, Jun.06, 2023. [Video]

Academic Service

Organizing Committee

Organizer, NeurIPS’26 Workshop on AI-Native Academia.
Organizer, TMLR Young Scientist Seminars.
Workshop Chair Assistant, NeurIPS 2024.
Workshop Assistant, HKBU-COMP & RIKEN-AIP Joint Workshop on AI and ML 2024.
Executive Group Member, RIKEN TrustML Young Scientist Seminars.
Conference Reviewer \ PC Member:

ICML [Top Reviewer, Area Chair], NeurIPS [Top Reviewer, Area Chair], ICLR [Notable Reviewer, Area Chair], AISTATS, KDD, ACML, AAAI, IJCAI.
Journal Reviewer:

ACM Computing Surveys, TPAMI, TMLR, IJCV, JAIR, TNNLS, MLJ, NN, PR.

Teaching

(Assistance in following courses)

COMP7250(PG): Machine Learning, Spring (2022)
COMP7240(PG): Recommender Systems, Autumn (2022, 2023)
COMP7160(PG): Research Methods in Computer Science, Autumn (2022,2023)
COMP4135(UG): Recommender Systems and Applications, Autumn (2022, 2023)
COMP3057(UG): Intro to AI and Machine Learning, Autumn (2022)