About Me


Welcome! I am Jiaxi Yang. I am a second-year Ph.D student, majoring in Informatics at the Pennsylvania State University (PSU). Previously, I got my master degree in 2024 from University of Electronic Science and Technology of China (UESTC).

My research interests broadly lie in Trustworthy AI, Federated Learning. I am also interested in Incentive Mechanism Design, Game Theory.

News

  • [08/2024] Starting my Ph.D at PSU! We are!!
  • [06/2024] One paper is accepted by IEEE Internet of Things Journal!
  • [06/2024] I am moving to Penn State University as a PhD student!
  • [05/2024] I successfully defended my master's thesis!
  • [10/2023] One paper is accepted by WCNC 2024!
  • [10/2023] Obtaining first-class scholarship at UESTC!
  • [06/2023] Our poster was presented in CVPR workshop!
  • [05/2023] Our paper is accepted by CVPR workshop!
  • [09/2022] Starting my research internship at UBC, Tuum est!
  • [07/2022] Our paper is accepted by WASA 2022!

    • Publications

    For the complete list, please check my google scholar. Email me if you have any questions about my papers or code, or if you would like to collaborate with me. (* indicates equal contribution)

  • Boosting E-commerce Content Diversity: A Graph-based RAG Approach with User Reviews
    Jiaxi Yang, Yiling Jia. Carl Yang Yi Liang Lu Lin KDD 2025
    [paper]
  • Unlearning for Federated Online Learning to Rank: A Reproducibility Study
    Yiling Tao, Shuyi Wang, Jiaxi Yang, Guido Zuccon. SIGIR 2025 (Reproducibility Paper)
    [paper]
  • GMValuator: Similarity-based Data Valuation for Generative Models.
    Jiaxi Yang*, Wenlong Deng*, Benlin Liu, Yangsibo Huang, James Zou, Xiaoxiao Li. ICLR 2025
    [paper]
  • Portfolio-Based Incentive Mechanism Design for Cross-Device Federated Learning.
    Jiaxi Yang, Sheng Cao, Cuifang Zhao, Weina Niu, Li-Chuan Tsai. WCNC 2024
    [paper]
  • Matching-based Data Valuation for Generative Model.
    Jiaxi Yang, Wenlong Deng, Benlin Liu, Yangsibo Huang, Xiaoxiao Li. CVPR workshop, 2023
    [paper]
  • Multi-dimensional Data Quick Query for Blockchain-based Federated Learning.
    Jiaxi Yang, Sheng Cao, Peng Xiangli, Xiong Li, Xiaosong Zhang. WASA 2022
    [paper] [video]
  • A Survey of Federated Unlearning: A Taxonomy, Challenges and Future Directions.
    Yang Zhao*, Jiaxi Yang*, Yiling Tao, Lixu Wang, Xiaoxiao Li, Dusit Niyato In submission
    [paper] [github]
  • TranDRL: A Transformer-Driven Deep Reinforcement Learning Enabled Prescriptive Maintenance Framework.
    Yang Zhao, Jiaxi Yang, Wenbo Wang, Helin Yang, Dusit Niyato IEEE Internet of Things Journal (Q1)
    [paper]
  • Federated Learning Incentive Mechanism under Buyers' Auction Market.
    Jiaxi Yang*, Zihao Guo*, Sheng Cao, Cuifang Zhao, Li-Chuan Tsai. In submission
    [paper]
  • A Survey on Trustworthiness in Foundation Models for Medical Image Analysis
    Congzhen Shi*, Ryan Rezai*, Jiaxi Yang*, Qi Dou, Xiaoxiao Li, In submission
    [paper]

    • Recent Projects

    Here are my recent exciting projects. For the full list of my projects, please refer to my publications.

    Portfolio-Based Incentive Mechanism Design for Cross-Device Federated Learning

    The paper proposes a new approach to designing incentive mechanisms for cross-device federated learning (FL) under real-world scenarios, taking into account the stochastic and hard-to-estimate nature of clients' performance. The proposed FL Incentive Mechanism is based on Portfolio theory and is the first application of this theory to resolve FL resource allocation problems. To accurately reflect practical FL scenarios, the paper introduces the Federated Learning Agent-Based Model (FL-ABM), which enables a deeper understanding of the factors that influence the system's outcomes.

    Matching-based Data Valuation for Generative Model

    The paper introduces Generative Model Valuator (GMValuator), a new approach for data valuation in deep generative models, which have been neglected by existing data valuation methods focused on discriminative models. GMValuator is model-agnostic in generative models, provides a training-free, post-hoc data valuation strategy, and is the first work to address the data valuation problem in deep generative models from a similarity-matching perspective.

    Multi-dimensional Data Quick Query for Blockchain-based Federated Learning

    The paper proposes a novel data structure called MerkleRB-Tree to improve query efficiency for multi-dimensional metadata in blockchain-based federated learning. Selecting suitable participants in FL by storing and querying metadata on blockchain could ensure reliable data owners and improve model quality.

    Patents

  • Insurance Compensation Apparatus and Method based on Hybrid Smart Contracts.
    Sheng Cao,Qian Zhang,Sixuan Dang, Guanghao Xiong,Haoqu Chen, Jiaxi Yang, Songdeng Niu, Zifeng Yang.
    CN202110835114.8
  • An Incentive Mechanism Method for Decision-Making .
    Sheng Cao, Sixuan Dang, Qian Zhang, Guanghao Xiong, Jiaxi Yang, Haoqu Chen, Songdeng Niu, Zifeng Yang.
    CN202110835181.X