Publications

Research on efficient inference, speculative decoding, and large language models.

My highlights

2026

  1. EAGLE-4: One Flexible Drafter for All Batch Sizes and Hardware

    EAGLE-4: One Flexible Drafter for All Batch Sizes and Hardware Highlighted by Jiale Fu

    EAGLE-4 Team
    Preprint 2026
    Core contributor to EAGLE-4; designed and implemented the Efficient LM Head and contributed additional architectural improvements that have not been publicly disclosed.
    Paper Blog Code
  2. Rethinking LLM Ensembling from the Perspective of Mixture Models

    Rethinking LLM Ensembling from the Perspective of Mixture Models Highlighted by Jiale Fu

    Jiale Fu, Yuchu Jiang, Peijun Wu, Chonghan Liu, Joey Tianyi Zhou, and Xu Yang
    ICML 2026 International Conference on Machine Learning
    Spotlight Paper Website
  3. GraphIC: A Graph-Based In-Context Example Retrieval Model for Multi-Step Reasoning

    GraphIC: A Graph-Based In-Context Example Retrieval Model for Multi-Step Reasoning

    Jiale Fu, Yaqing Wang, Simeng Han, Jiaming Fan, and Xu Yang
    AAAI 2026 AAAI Conference on Artificial Intelligence
    Paper Website
  4. TreeGraft: Adaptive Multi-Drafter Grafting for Tree-Based Speculative Decoding

    TreeGraft: Adaptive Multi-Drafter Grafting for Tree-Based Speculative Decoding

    Jiaming Fan*, Daming Cao*, Canchen Huang, Jiale Fu, Jin Zhang, Junjie Gao, Kai Yang, Xiangzhong Luo, and Xu Yang
    NeurIPS 2026 Advances in Neural Information Processing Systems
    Spotlight Paper Website
  5. Flatter Tokens are More Valuable for Speculative Draft Model Training

    Flatter Tokens are More Valuable for Speculative Draft Model Training

    Jiaming Fan, Daming Cao, Xiangzhong Luo, Jiale Fu, Chonghan Liu, and Xu Yang
    ICLR 2026 International Conference on Learning Representations
    Paper Code Website
  6. d^2Cache: Accelerating Diffusion-Based LLMs via Dual Adaptive Caching

    d^2Cache: Accelerating Diffusion-Based LLMs via Dual Adaptive Caching

    Yuchu Jiang, Yue Cai, Xiangzhong Luo, Jiale Fu, Jiarui Wang, Chonghan Liu, and Xu Yang
    ICLR 2026 International Conference on Learning Representations
    Paper Code Website

2025

  1. Fast Large Language Model Collaborative Decoding via Speculation

    Fast Large Language Model Collaborative Decoding via Speculation Highlighted by Jiale Fu

    Jiale Fu*, Yuchu Jiang*, Junkai Chen, Jiaming Fan, Xin Geng, and Xu Yang
    ICML 2025 International Conference on Machine Learning
    Paper Video Blog Code Website
  2. Mimic In-Context Learning for Multimodal Tasks

    Mimic In-Context Learning for Multimodal Tasks

    Yuchu Jiang, Jiale Fu, Chenduo Hao, Xinting Hu, Yingzhe Peng, Xin Geng, and Xu Yang
    CVPR 2025 IEEE/CVF Conference on Computer Vision and Pattern Recognition
    Paper Blog Code Website

2023

  1. Identifying vital nodes in recovering dynamical process of networked system

    Identifying vital nodes in recovering dynamical process of networked system

    Jiale Fu, Xiaoya Jiang, Qi Shao, Duxin Chen, and Wenwu Yu
    Frontiers 2023 Frontiers in Physics
    Paper Website