Ph.D. Student Computer Science & Engineering
I study reliable and aligned LLM agents, with an emphasis on failure diagnosis, deployment-time risks, and trustworthy evaluation.
Currently diagnosing how LLM agents break down under input faults
I am a first-year Ph.D. student at the University of Notre Dame, advised by Prof. Fanny YePh.D. advisor, Department of Computer Science & Engineering, since Sep 2026.. Prior to that, I completed my undergraduate studies in Cyber Science and Engineering at Sichuan UniversityB.Eng., School of Cyber Science and Engineering, Chengdu, 2021–2026..
During my visiting studies, I was mentored by Prof. Xiangliang ZhangVisiting research student at Notre Dame since Aug 2024, working on trustworthy generative models and agent evaluation. and senior researcher Yue Huang, whose guidance shaped my research interests in trustworthy AI and foundation models. I continue to collaborate closely with Zheyuan Zhang.
My research focuses on trustworthy AI for LLM agents, with an emphasis on reliability, alignment, and evaluation under realistic training and deployment conditions.
One paper was accepted to the EMNLP 2026 Main Conference.
AutoDavis was accepted to the KDD 2026 Datasets & Benchmarks Track. Two papers were accepted to ICML 2026: Alignment Risks from Capability-Seeking RL Training and Drift-Bench.
Two papers were accepted to ACL 2026 Findings.
One paper was accepted to WWW 2026 Demo Track. One paper was accepted to ICLR 2026.
One paper was accepted to AAAI 2025.
I make LLM agents dependable: diagnosing how they fail, keeping them aligned as capabilities and deployments shift, and measuring whether they can be trusted.
01 Agent Reliability
I study how LLM agents fail across multi-turn interaction, input faults, and cooperative settings, how those failures can be diagnosed, and how execution experience can make agent harnesses more adaptive.
02 Robust Alignment
I investigate how alignment risks emerge as models acquire new capabilities or move into constrained deployment environments, together with guardian and advisor mechanisms for safer behavior.
03 Trustworthy Evaluation
I develop benchmarks and evaluation protocols that test whether generative, vision-language, and domain-specific foundation models remain reliable in realistic use cases.
* Equal contribution. See also Google Scholar.
ICML 2026 International Conference on Machine Learning
@inproceedings{bao2026drift,
title = {{Drift-Bench: Diagnosing CoopeRative Breakdowns in LLM Agents under Input Faults via Multi-Turn Interaction}},
author = {Han Bao and Zheyuan Zhang and Pengcheng Jing and Zhengqing Yuan and Kaiwen Shi and Yanfang Ye},
booktitle = {International Conference on Machine Learning},
year = {2026}
} ICML 2026 International Conference on Machine Learning
@inproceedings{zhou2026alignment,
title = {{Alignment Risks from Capability-Seeking RL Training}},
author = {Yujun Zhou and Yue Huang and Han Bao and Kehan Guo and Zhenwen Liang and Pin-Yu Chen and Tian Gao and Werner Geyer and Nuno Moniz and Nitesh V Chawla and Xiangliang Zhang},
booktitle = {International Conference on Machine Learning},
year = {2026}
} arXiv 2026
@article{huang2026memoharness,
title = {{MemoHarness: Agent Harnesses That Learn from Experience}},
author = {Yue Huang and Wenjie Wang and Han Bao and Yuchen Ma and Xiaonan Luo and Yi Nian and Haomin Zhuang and Zheyuan Liu and Yue Zhao and Xiangliang Zhang},
journal = {arXiv preprint arXiv:2607.14159},
year = {2026}
} arXiv 2026 Position paper
@article{bao2026ai,
title = {{AI Alignment Breaks at the Edge}},
author = {Han Bao and Yue Huang and Xiaoda Wang and Zheyuan Zhang and Yujun Zhou and Carl Yang and Xiangliang Zhang and Yanfang Ye},
journal = {arXiv preprint arXiv:2602.20042},
year = {2026}
} KDD 2026 ACM SIGKDD Conference on Knowledge Discovery and Data Mining, Datasets and Benchmarks Track
@inproceedings{bao2026autodavis,
title = {{AutoDavis: Automatic Benchmarking of Large Vision-Language Models on Visual Question-Answering}},
author = {Han Bao and Yue Huang and Yanbo Wang and Jiayi Ye and Xiangqi Wang and Xiuying Chen and Mohamed Elhoseiny and Xiangliang Zhang},
booktitle = {ACM SIGKDD Conference on Knowledge Discovery and Data Mining, Datasets and Benchmarks Track},
year = {2026}
} ACL Findings 2026 Findings of the Association for Computational Linguistics
@inproceedings{bao2026policyllm,
title = {{PolicyLLM: Towards Excellent Comprehension of Public Policy for Large Language Models}},
author = {Han Bao and Penghao Zhang and Yue Huang and Rui Su and Zhengqing Yuan and Yujun Zhou and Xiangqi Wang and Kehan Guo and Nitesh Chawla and Xiangliang Zhang},
booktitle = {Findings of the Association for Computational Linguistics},
year = {2026}
} WWW 2026 Demo The ACM Web Conference, Demo Track
@inproceedings{bao2026intraai,
title = {{IntraAI: Bridging Human-AI Understanding Through Intelligent Interface Design}},
author = {Han Bao and Yue Huang and Xiangliang Zhang and Yanfang Ye},
booktitle = {The ACM Web Conference, Demo Track},
year = {2026}
} EMNLP 2026 Conference on Empirical Methods in Natural Language Processing
@inproceedings{wang2026specalign,
title = {{SpecAlign: Efficient Specification-Grounded Alignment of Large Language Models via Synthetic Data}},
author = {Wenjie Wang and Yue Huang and Zhengqing Yuan and Han Bao and Shiyi Du and Yuchen Ma and Yue Zhao and Yanfang Ye and Xiangliang Zhang},
booktitle = {Conference on Empirical Methods in Natural Language Processing},
year = {2026}
} ACL Findings 2026 Findings of the Association for Computational Linguistics
@inproceedings{huang2026guardian,
title = {{Guardian-as-an-Advisor: Advancing Next-Generation Guardian Models for Trustworthy LLMs}},
author = {Yue Huang and Haomin Zhuang and Jiayi Ye and Han Bao and Yanbo Wang and Hang Hua and Siyuan Wu and Pin-Yu Chen and Xiangliang Zhang},
booktitle = {Findings of the Association for Computational Linguistics},
year = {2026}
} ICLR 2026 International Conference on Learning Representations Leader of T2I Group
@inproceedings{huang2026trustworthiness,
title = {{On the Trustworthiness of Generative Foundation Models: Guideline, Assessment, and Perspective}},
author = {Yue Huang and Chujie Gao and Siyuan Wu and Haomin Zhuang and Jiayi Ye and Han Bao and others},
booktitle = {International Conference on Learning Representations},
year = {2026}
} arXiv 2026
@article{ma2026skillgen,
title = {{SkillGen: Verified Inference-Time Agent Skill Synthesis}},
author = {Yuchen Ma and Yue Huang and Han Bao and Haomin Zhuang and Swadheen Shukla and Michel Galley and Xiangliang Zhang and Stefan Feuerriegel},
journal = {arXiv preprint arXiv:2605.10999},
year = {2026}
} arXiv 2026
@article{shi2026confidence,
title = {{Confidence Laundering in Agent Systems: Why Uncertainty Needs a Latent Carrier}},
author = {Kaiwen Shi and Zheyuan Zhang and Han Bao and Colby Nelson and Yanfang Ye},
journal = {arXiv preprint arXiv:2606.20662},
year = {2026}
} arXiv 2026
@article{wang2026uxbench,
title = {{UXBench: Measuring the Actionability of LLM-Generated UX Critiques}},
author = {Wenjie Wang and Yue Huang and Zipeng Ling and Han Bao and Hang Hua and Xiaonan Luo and Yu Jiang and Shiyi Du and Yuexing Hao and Xiaomin Li and Yuchen Ma and Dianzhuo Wang and Yanfang Ye and Xiangliang Zhang},
journal = {arXiv preprint arXiv:2606.16262},
year = {2026}
} arXiv 2026
@article{zhang2026semantic,
title = {{Why Semantic Entropy Fails? Geometry-Aware and Calibrated Uncertainty for Policy Optimization}},
author = {Zheyuan Zhang and Kaiwen Shi and Han Bao and Zehong Wang and Tianyi Ma and Yanfang Ye},
journal = {arXiv preprint arXiv:2605.21801},
year = {2026}
} AAAI 2025 AAAI Conference on Artificial Intelligence
@inproceedings{liu2025lid,
title = {{LiD-FL: Towards List-Decodable Federated Learning}},
author = {Hong Liu and Liren Shan and Han Bao and Ronghui You and Yuhao Yi and Jiancheng Lv},
booktitle = {AAAI Conference on Artificial Intelligence},
year = {2025}
} BMC MIDM 2025 BMC Medical Informatics and Decision Making
@article{wang2025crisp,
title = {{CRISP: A Causal Relationships-Guided Deep Learning Framework for Advanced ICU Mortality Prediction}},
author = {Linna Wang and Xinyu Guo and Han Bao and Lihua Jiang and Li Zhao and Tao Zhu and Ziliang Feng},
journal = {BMC Medical Informatics and Decision Making},
year = {2025}
} University of Notre Dame Sep 2026 – Present
Ph.D. in Computer Science · Advisor: Prof. Fanny Ye
Sichuan University Sep 2021 – Jun 2026
B.Eng. in Cyber Science and Engineering
University of Notre Dame, Visiting Research Student Aug 2024 – Present
Advisor: Prof. Xiangliang Zhang · Trustworthy Generative Models, LLM Alignment, Agent Evaluation
National University of Singapore, Summer Workshop Participant Jul 2024
Advisor: Prof. Tianbai Ma · Cloud Computing, Network Traffic Analysis
Provincial Third Prize, China Undergraduate Mathematical Contest in Modeling Nov 2023
Chinese Society for Industrial and Applied Mathematics (CSIAM)
Second-Class Scholarship (Top 10%) Sep 2023
School of Cyber Science and Engineering, Sichuan University
Reviewer 2027
International Conference on Learning Representations (ICLR 2027)
Reviewer 2027
The ACM Web Conference (WWW 2027)
Reviewer 2026
AAAI Conference on Artificial Intelligence (AAAI 2026)
Program Committee / Reviewer 2025
NeurIPS 2025 Workshop on Socially Responsible and Trustworthy Foundation Models