Yuxin Tian

me.jpeg

My name is Yuxin Tian. I obtained my Ph.D. in Computer Science and Technology from Sichuan University, where I was advised by Prof. Jiancheng Lv and closely collaborated with Prof. Xi Peng. Before my graduate studies, I also received my B.S. from the Sichuan University. My research interests focus on Large Language Model (LLM) pre-training, efficient model architecture, and robust learning in open scenarios, with a focus on Multi-modal Learning, Mixture-of-expert, Multi-task Learning, and Learning with Noisy Labels.

Currently, I am a researcher at the pre-training team of inclusionAI, Ant Group.

Service:

  • Conference: Reviewer of ICLR, NeurIPS, ICML, CVPR, ICCV, ECCV, AISTATS, AAAI, IJCAI, ACM MM, and etc.
  • Journal: Reviewer of IEEE-TNNLS, IEEE-TCYB, IEEE-TSMC:Systems, and IEEE-TCE

News

Jun 16, 2026 The techniqual report of Ling-2.6 & Ring-2.6 is announced.
May 01, 2026 One first-authored paper Learning Locally, Revising Globally: Global Reviser for Federated Learning with Noisy Labels was accepted by Forty-Third International Conference on Machine Learning (ICML 2026). Congrats to all authors!
Feb 22, 2026 One co-authored paper HyperNAS: Enhancing Architecture Representation for NAS Predictor via Hypernetwork have been accepted by The IEEE/CVF Conference on Computer Vision and Pattern Recognition 2026 (CVPR 2026). Congrats to all my co-authors!
Jan 26, 2026 One co-authored paper Deploying Models to Non-participating Clients in Federated Learning without Fine-tuning: A Hypernetwork-based Approach have been accepted by The Fourteenth International Conference on Learning Representations (ICLR 2026). Congrats to all my co-authors!
Feb 27, 2025 Two co-authored papers Style Quantization for Data-Efficient GAN Training and Ferret: An Efficient Online Continual Learning Framework under Varying Memory Constraints have been accepted by Conference on Computer Vision and Pattern Recognition 2025 (CVPR 2025). Congrats to all my co-authors!

Selected Publications

  1. Learning Locally, Revising Globally: Global Reviser for Federated Learning with Noisy Labels
    Yuxin Tian, Mouxing Yang, Yuhao Zhou, and 5 more authors
    2026
  2. Ling and Ring 2.6 Technical Report: Efficient and Instant Agentic Intelligence at Trillion-Parameter Scale
    Ang Li, Ben Liu, Bin Han, and 215 more authors
    2026
  3. E-3SFC: Communication-Efficient Federated Learning With Double-Way Features Synthesizing
    Yuhao Zhou, Yuxin Tian, Mingjia Shi, and 4 more authors
    IEEE Transactions on Neural Networks and Learning Systems, 2025
    DOI
  4. An Empirical Study of Parameter Efficient Fine-tuning on Vision-Language Pre-train Model
    Yuxin Tian, Mouxing Yang, Yunfan Li, and 4 more authors
    In IEEE International Conference on Multimedia and Expo, ICME 2024, Niagara Falls, ON, Canada, July 15-19, 2024, 2024
    DOI
  5. UNITE: Multitask Learning With Sufficient Feature for Dense Prediction
    Yuxin Tian, Yijie Lin, Qing Ye, and 3 more authors
    IEEE Trans. Syst. Man Cybern. Syst., 2024
    DOI