• About me

    Short Bio. I am a tenure-track Assistant Professor at the Department of Computer Science with a joint appointment in the School of Data Science and Society, University of North Carolina at Chapel Hill. My lab, AIMING, studies adaptive intelligence through alignment, interaction and learning, and is affiliated with the UNC NLP group. I was a Postdoctoral Scholar at Stanford University hosted by Chelsea Finn. I received my Ph.D. degree in 2021 at Pennsylvania State University under the advisory of Zhenhui (Jessie) Li. During my Ph.D, I also spent time visiting CMU hosted by Eric P. Xing.

     

    Lab Openings. We will recruit 2-3 Ph.D. students for Fall 2025 and multiple interns or visiting students all year. Please read THIS for detailed recruitment information.

     

    Research Interests. My research focuses on both the theoretical and applied aspects of building generalizable and reliable foundation models (e.g., LLMs, VLMs, Diffusion Models). Additionally, I am keen on utilizing these models to facilitate applications in robotics and biomedicine. Currently, my primary endeavors revolve around the following key directions:

    1. Developing automated mechanisms to identify potential failures in the generalizability of foundation models (e.g., hallucination, bias).
    2. Exploring efficient strategies to enable AI models to overcome reliability challenges and generalize to diverse environments.
    3. Steering AI models towards better alignment with human objectives, preferences, and ethical values.

    Related ML topics: AI alignment and preference learning (e.g., RLHF), out-of-distribution generalization, hallucination detection and mitigation, meta-learning and continual learning, fairness and bias, uncertainty estimation and calibration.

    Focused applications: robot learning and embodied AI, biomedicine and healthcare.

     

    You can follow me on Twitter at @HuaxiuYaoML or 小红书 at Huaxiu Yao.

    News and Travel

    [2024-2025 Service] Area Chair in ICML 2024, NeurIPS 2024, ICLR 2025, ACL Rolling Review, AISTATS 2025

    [2024.09] Five papers were accepted by NeurIPS 2024 (three main track, two D&B track), One paper was accepted by EMNLP 2024

    [2024.08] Won the KDD 2024 Best Paper Award

    [2024.07] One paper was accepted by ECCV 2024, one paper was accepted by COLM 2024

    [2024.05] One paper was accepted by ACL 2024, one paper was accepted by TMLR

    [2024.05] Five papers were accepted by ICML 2024 (four regular papers, one position paper)

    [2024.04] We will organize Workshop on Foundation Models in the Wild at ICML 2024. Stay tuned!

    [2024.03] One paper was accepted by NAACL 2024

    [2024.02] One paper was accepted by CVPR 2024

    [2024.01] Three papers were accepted by ICLR 2024

    Awards

    • KDD Best Paper Award, 2024
    • Cisco Faculty Research Award, 2024
    • Creativity Hubs Seed-funding Winner, 2024
    • Google Cloud Research Credits Award, 2024
    • NC TraCS Pilot Award, 2024
    • AAAI New Faculty Highlights, 2024
    • AI2000 Most Influential Scholar Award Honorable Mention, 2022
    • AI Rising Stars in Chinese Students, Baidu Research, 2021
    • College of IST Ph.D. Award for Research Excellence, Penn State University, 2020
  • publicationS

    Here are selected publications that most closely align with my research focus. Please see the complete list here or Google Scholar.

    The underline authors are students (co-)mentored by me; : equal advising

    [AI Alignment], [ML Generalization], [Embodied Intelligence], [Multimodal AI], [AI for Health]

    Preprints

    [1] Peng Xia, Kangyu Zhu, Haoran Li, Tianze Wang, Weijia Shi, Sheng Wang, Linjun Zhang, James Zou, Huaxiu Yao, MMed-RAG: Versatile Multimodal RAG System for Medical Vision Language Models, arXiv 2410.13085. [arXiv] [Code]

    [AI Alignment], [Multimodal AI], [AI for Health]

     

    [2] Zhaoyang Wang, Weilei He, Zhiyuan Liang, Xuchao Zhang, Chetan Bansal, Ying Wei, Weitong Zhang, Huaxiu Yao, CREAM: Consistency Regularized Self-Rewarding Language Models, arXiv 2410.12735. [arXiv] [Code]

    [AI Alignment]

     

    [3] Peng Xia*, Siwei Han*, Shi Qiu*, Yiyang Zhou, Zhaoyang Wang, Wenhao Zheng, Zhaorun Chen, Chenhang Cui, Mingyu Ding, Linjie Li, Lijuan Wang, Huaxiu Yao, MMIE: Massive Multimodal Interleaved Comprehension Benchmark for Large Vision-Language Models, arXiv 2410.10139. [arXiv] [Project Page]

    [Multimodal AI]

     

    [4] Zhaorun Chen*, Yichao Du*, Zichen Wen*, Yiyang Zhou*, Chenhang Cui, Zhenzhen Weng, Haoqin Tu, Chaoqi Wang, Zhengwei Tong, Qinglan Huang, Canyu Chen, Qinghao Ye, Zhihong Zhu, Yuqing Zhang, Jiawei Zhou, Zhuokai Zhao, Rafael Rafailov, Chelsea Finn, Huaxiu Yao, MJ-Bench: Is Your Multimodal Reward Model Really a Good Judge for Text-to-Image Generation?, arXiv 2407.04842. [arXiv] [Project Page]

    [AI Alignment], [Multimodal AI]

     

    [5] Yiyang Zhou*, Chenhang Cui*, Rafael Rafailov, Chelsea Finn, Huaxiu Yao, Aligning Modalities in Vision Large Language Models via Preference Fine-tuning, arXiv 2402.11411. [arXiv] [Code]

    [AI Alignment], [Multimodal AI]

     

    [6] Taiming Lu*, Lingfeng Shen*, Xinyu Yang*, Weiting Tan, Beidi Chen, Huaxiu Yao, It Takes Two: On the Seamlessness between Reward and Policy Model in RLHF, arXiv 2406.07971. [arXiv]

    [AI Alignment]

     

    [7] Chenhang Cui*, Yiyang Zhou*, Xinyu Yang, Shirley Wu, Linjun Zhang, James Zou, Huaxiu Yao, Holistic Analysis of Hallucination in GPT-4V(ision): Bias and Interference Challenges, arXiv 2311.03287. [arXiv] [Code]

    [Multimodal AI], [ML Generalization]

    2024

    [1] Yiyang Zhou*, Zhiyuan Fan*, Dongjie Cheng*, Sihan Yang, Zhaorun Chen, Chenhang Cui, Xiyao Wang, Yun Li, Linjun Zhang, Huaxiu Yao, Calibrated Self-Rewarding Vision Language Models, in Proceeding of the Thirty-Eighth Conference on Neural Information Processing Systems (NeurIPS 2024), Vancouver, Canada, Dec 2024. [arXiv] [Code]

    [AI Alignment], [Multimodal AI]

     

    [2] Xinyu Yang, Jixuan Leng, Geyang Guo, Jiawei Zhao, Ryumei Nakada, Linjun Zhang, Huaxiu Yao, Beidi Chen, S^2FT: Efficient, Scalable and Generalizable LLM Fine-tuning by Structured Sparsity, in Proceeding of the Thirty-Eighth Conference on Neural Information Processing Systems (NeurIPS 2024), Vancouver, Canada, Dec 2024.

    [ML Generalization]

     

    [3] Peng Xia, Ze Chen, Juanxi Tian, Yangrui Gong, Ruibo Hou, Yue Xu, Zhenbang Wu, Zhiyuan Fan, Yiyang Zhou, Kangyu Zhu, Wenhao Zheng, Zhaoyang Wang, Xiao Wang, Xuchao Zhang, Chetan Bansal, Marc Niethammer, Junzhou Huang, Hongtu Zhu, Yun Li, Jimeng Sun, Zongyuan Ge, Gang Li, James Zou, Huaxiu Yao, CARES: A Comprehensive Benchmark of Trustworthiness in Medical Vision Language Models, in Proceeding of the Thirty-Eighth Conference on Neural Information Processing Systems Track on  Datasets & Benchmarks (NeurIPS 2024), Vancouver, Canada, Dec 2024. [arXiv] [Code]

    [Multimodal AI], [AI for Health]

     

    [4] Tony Lee*, Haoqin Tu*, Chi Heem Wong*, Wenhao Zheng, Yiyang Zhou, Yifan Mai, Josselin Somerville Roberts, Michihiro Yasunaga, Huaxiu Yao, Cihang Xie, Percy Liang, VHELM: A Holistic Evaluation of Vision Language Models, in Proceeding of the Thirty-Eighth Conference on Neural Information Processing Systems Track on  Datasets & Benchmarks (NeurIPS 2024), Vancouver, Canada, Dec 2024.

    [AI Alignment], [ML Generalization], [Multimodal AI]

     

    [5] Zhen-Yu Zhang, Zhiyu Xie, Huaxiu Yao, Masashi Sugiyama, Test-time Adaptation in Non-stationary Environments via Adaptive Representation Alignment, in Proceeding of the Thirty-Eighth Conference on Neural Information Processing Systems (NeurIPS 2024), Vancouver, Canada, Dec 2024.

    [AI Alignment], [ML Generalization]

     

    [6] Peng Xia*, Kangyu Zhu*, Haoran Li, Hongtu Zhu, Yun Li, Gang Li, Linjun Zhang, Huaxiu Yao, RULE: Reliable Multimodal RAG for Factuality in Medical Vision Language Models, in Proceedings of the 2024 Conference on Empirical Methods in Natural Language Processing (EMNLP 2024), Miami, Nov. 2024. [arXiv] [Code]

    [AI Alignment], [Multimodal AI], [AI for Health]

     

    [7] Xiyao Wang, Yuhang Zhou, Xiaoyu Liu, Hongjin Lu, Yuancheng Xu, Feihong He, Jaehong Yoon, Taixi Lu, Gedas Bertasius, Mohit Bansal, Huaxiu Yao, Furong Huang, Mementos: A Comprehensive Benchmark for Multimodal Large Language Model Reasoning over Image Sequences, in Proceeding of the 62nd Annual Meeting of the Association for Computational Linguistics (ACL 2024), Bangkok, Thailand, Aug 2024. [arXiv] [Code]

    [Multimodal AI]

     

    [8] Haoqin Tu*, Chenhang Cui*, Zijun Wang*, Yiyang Zhou, Bingchen Zhao, Junlin Han, Wangchunshu Zhou, Huaxiu Yao, Cihang Xie, How Many Unicorns Are in This Image? A Safety Evaluation Benchmark for Vision LLMs, in Proceeding of the 18th European Conference on Computer Vision (ECCV 2024), Milan, Italy, Sept 2024. [arXiv] [Code]

    [AI Alignment], [ML Generalization], [Multimodal AI]

     

    [9] Zhaorun Chen, Zhuokai Zhao, Hongyin Luo, Huaxiu Yao, Bo Li, Jiawei Zhou, HALC: Object Hallucination Reduction via Adaptive Focal-Contrast Decoding, in Proceeding of the Forty-first International Conference on Machine Learning (ICML 2024), Vienna, Austria, July 2024. [Paper] [Code]

    [AI Alignment], [Multimodal AI]

     

    [10] Xu Yang, Huaxiu Yao, Ying Wei, One Meta-tuned Transformer is What You Need for Few-shot Learning, in Proceeding of the Forty-first International Conference on Machine Learning (ICML 2024), Vienna, Austria, July 2024 (Spotlight).

    [ML Generalization]

     

    [11] Yiyang Zhou*, Chenhang Cui*, Jaehong Yoon, Linjun Zhang, Zhun Deng, Chelsea Finn, Mohit Bansal, Huaxiu Yao, Analyzing and Mitigating Object Hallucination in Large Vision-Language Models, in Proceeding of the 12th International Conference on Learning Representations (ICLR 2024), Vienna, Austria, May 2024  (the short version is presented in NeurIPS 2023 Instruction Workshop). [Paper] [Code]

    [AI Alignment], [ML Generalization]

     

    [12] Huaxiu Yao*, Xinyu Yang*, Xinyi Pan, Shengchao Liu, Pang Wei Koh, Chelsea Finn, Improving Domain Generalization with Domain Relations, in Proceeding of the 12th International Conference on Learning Representations (ICLR 2024), Vienna, Austria, May 2024 (Spotlight). [Paper]

    [AI Alignment], [ML Generalization]

     

    [13] Katherine Tian*, Eric Mitchell*, Huaxiu Yao, Christopher D Manning, Chelsea Finn, Fine-Tuning Language Models for Factuality, in Proceeding of the 12th International Conference on Learning Representations (ICLR 2024), Vienna, Austria, May 2024  (the short version is presented in NeurIPS 2023 Instruction Workshop). [Paper]

    [AI Alignment]

     

    [14] Xiaohui Zhang, Jaehong Yoon, Bansal Mohit, Huaxiu Yao, Multimodal Representation Learning by Alternating Unimodal Adaptation, in Proceeding of the IEEE/CVF Conference on Computer Vision and Pattern Recognition 2024 (CVPR 2024), Seattle, WA, Jun 2024. [Paper]

    [Multimodal AI]

     

    [15] Caroline Choi*, Fahim Tajwar*, Yoonho Lee*, Huaxiu Yao, Ananya Kumar, Chelsea Finn, Conservative Prediction via Data-Driven Confidence Minimization, Transactions on Machine Learning Research (TMLR), 2024. [Paper]

    [ML Generalization]

    2023

    [16] Percy Liang, Rishi Bommasani, Tony Lee, [and 47 others, including Huaxiu Yao], Holistic Evaluation of Language Models, Transactions on Machine Learning Research (TMLR, Featured), 2023. [Paper] [Project Page]

    [AI Alignment]

     

    [17] Katherine Tian*, Eric Mitchell*, Allan Zhou, Archit Sharma, Rafael Rafailov, Huaxiu Yao, Chelsea Finn, Christopher D Manning, Just Ask for Calibration: Strategies for Eliciting Calibrated Confidence Scores from Language Models Fine-Tuned with Human Feedback, in Proceedings of the 2023 Conference on Empirical Methods in Natural Language Processing (EMNLP 2023), Singapore, Dec. 2023 (Short Paper). [Paper]

    [AI Alignment]

     

    [18] Zhenbang Wu, Huaxiu Yao, David Liebovitz, Jimeng Sun, An Iterative Self-Learning Framework for Medical Domain Generalization, in Proceeding of the Thirty-Seventh Conference on Neural Information Processing Systems (NeurIPS 2023), New Orleans, LA, Dec. 2023. [Paper]

    [ML Generalization], [AI for Health]

     

    [19] Yoonho Lee, Huaxiu Yao, Chelsea Finn, Diversify and Disambiguate: Learning From Underspecified Data, in Proceeding of the 11th International Conference on Learning Representations (ICLR 2023), Kigali, Rwanda, May 2023 (the short version is presented in ICML 2022 Workshop on Spurious correlations, Invariance, and Stability and Workshop on Principles of Distribution Shift). [Paper] [Code]

    [ML Generalization]

     

    [20] Yoonho Lee*, Annie S. Chen*, Fahim Tajwar, Ananya Kumar, Huaxiu Yao, Percy Liang, Chelsea Finn, Surgical Fine-Tuning Improves Adaptation to Distribution Shifts, in Proceeding of the 11th International Conference on Learning Representations (ICLR 2023), Kigali, Rwanda, May 2023 (the short version is presented in NeurIPS 2022 I Can't Believe It's Not Better Workshop and Workshop on Distribution Shifts). [Paper] [Code]

    [AI Alignment], [ML Generalization]

     

    [21] Xinzhe Zuo, Zixiang Chen, Huaxiu Yao, Yuan Cao, Quanquan Gu, Understanding Train-Validation Split in Meta Learning with Neural Networks, in Proceeding of the 11th International Conference on Learning Representations (ICLR 2023), Kigali, Rwanda, May 2023. [Paper]

    [ML Generalization]

    2022

    [22] Huaxiu Yao*, Yiping Wang*, Linjun Zhang, James Zou, Chelsea Finn, C-Mixup: Improving Generalization in Regression, in Proceeding of the Thirty-Sixth Conference on Neural Information Processing Systems (NeurIPS 2022), New Orleans, LA, Dec. 2022. [Paper] [Code] [Video] [Bilibili] [Slides]

    [ML Generalization]

     

    [23] Huaxiu Yao*, Caroline Choi*, Bochuan Cao, Yoonho Lee, Pang Wei Koh, Chelsea Finn, Wild-Time: A Benchmark of in-the-Wild Distribution Shift over Time, in Proceeding of the Thirty-Sixth Conference on Neural Information Processing Systems Track on  Datasets & Benchmarks (NeurIPS 2022), New Orleans, LA, Dec. 2022. [Paper] [Code] [Website] [Video] [Bilibili] [Slides]

    [ML Generalization]

     

    [24] Huaxiu Yao*, Yu Wang*, Sai Li, Linjun Zhang, Weixin Liang, James Zou, Chelsea Finn, Improving Out-of-Distribution Robustness via Selective Augmentation, in Proceeding of the Thirty-ninth International Conference on Machine Learning (ICML 2022), Baltimore, MD, July 2022. [Paper] [Code] [Video] [Bilibili] [Slides]

    [ML Generalization]

     

    [25] Huaxiu Yao, Linjun Zhang, Chelsea Finn, Meta-Learning with Fewer Tasks through Task Interpolation, in Proceeding of the 10th International Conference on Learning Representations (ICLR 2022), Virtual, Apr. 2022 (Oral, 54/3391). [Paper] [Code] [Slides]

    [ML Generalization]

    Before 2022

    [26] Huaxiu Yao*, Yu Wang*, Ying Wei, Peilin Zhao, Mehrdad Mahdavi, Defu Lian, Chelsea Finn, Meta-learning with an Adaptive Task Scheduler, in Proceeding of the Thirty-fifth Conference on Neural Information Processing Systems (NeurIPS 2021), Virtual, Dec. 2021. [Paper] [Code] [Slides]

    [ML Generalization]

     

    [27] Huaxiu Yao, Ying Wei, Longkai Huang, Ding Xue, Junzhou Huang, Zhenhui Li, Functionally Regionalized Knowledge Transfer for Low-resource Drug Discovery, in Proceeding of the Thirty-fifth Conference on Neural Information Processing Systems (NeurIPS 2021), Virtual, Dec. 2021. [Paper][Code] [Slides]

    [ML Generalization], [AI for Health]

     

    [28] Huaxiu Yao*, Longkai Huang*, Linjun Zhang, Ying Wei, Li Tian, James Zou, Junzhou Huang, Zhenhui Li, Improving Generalization in Meta-learning via Task Augmentation, in Proceeding of the Thirty-eighth International Conference on Machine Learning (ICML 2021), Virtual, Jul. 2021. [Paper] [Code] [Slides]

    [ML Generalization]

     

    [29] Huaxiu Yao, Yingbo Zhou, Mehrdad Mahdavi, Zhenhui Li, Richard Socher, Caiming Xiong, Online Structured Meta-learning, in Proceeding of the Thirty-fourth Conference on Neural Information Processing Systems (NeurIPS 2020), Virtual, Dec. 2020. [Paper] [Slides]

    [ML Generalization]

     

    [30] Huaxiu Yao, Xian Wu, Zhiqiang Tao, Yaliang Li, Bolin Ding, Ruirui Li, Zhenhui Li, Automated Relational Meta-learning, in Proceeding of the Eighth International Conference on Learning Representations (ICLR 2020), Virtual, Apr. 2020. [Paper] [Code] [Slides]

    [ML Generalization]

     

    [31] Huaxiu Yao, Ying Wei, Junzhou Huang, Zhenhui Li, Hierarchically Structured Meta-learning, in Proceeding of the 36th International Conference on Machine Learning (ICML 2019), Long Beach, CA, June 2019. [Paper] [Code] [Slides]

    [ML Generalization]

  • Teaching

    Lecture

    • CS 590/790-183: Transfer Learning, UNC-CH, Spring 2024
    • CS 790-150: Reliable Machine Learning, UNC-CH, Fall 2023, Fall 2024
    • CS 330: Deep Multi-Task and Meta Learning (Domain Generalization), Stanford University, Fall 2022

    Tutorial

    • Learning with Small Data. (KDD 2020 [Website] [Slides] [YouTube] [Bilibili]) (WSDM 2020 [Website]) (AAAI 2021)
    • Meta-learning and Automated Machine Learning: Approaches and Applications​. (IJCAI 2020)

  • Service

    Conference Area Chair

    • International Conference on Machine Learning (ICML), 2024
    • Conference on Neural Information Processing Systems (NeurIPS), 2024; D&B Track (2022 - 2024)
    • International Conference on Learning Representations (ICLR), 2025
    • International Conference on Artificial Intelligence and Statistics (AISTATS), 2024
    • Empirical Methods in Natural Language Processing  (EMNLP), 2024
    • International Conference on Automated Machine Learning (AutoML-Conf), 2022 - 2024
    • Learning on Graphs Conference (LoG), 2022 - 2024

    Workshop Organizer

  • contact

    Office: 254, Sitterson Hall, Chapel Hill, NC 27599