publications

publications by categories in reversed chronological order.

2027

  1. ASPLOS
    Axe: A Simple Unified Layout Abstraction for Machine Learning Compilers
    Bohan Hou, Hongyi Jin, Guanjie Wang, Jinqi Chen, Yaxing Cai, Lijie Yang, Zihao Ye, Yaoyao Ding, Ruihang Lai, and 1 more author
    In Proceedings of the International Conference on Architectural Support for Programming Languages and Operating Systems, 2027
    Website

2026

  1. arXiv
    AbridgeKV: Compressing Long Reasoning into the KV Cache of Learned Pause Tokens
    Jack Cai, Lijie Yang, Vasilis Kontonis, Dimitris Papailiopoulos, and Shivam Garg
    2026
    Website
  2. COLM
    Thought-Level Beam Search for Reasoning
    Lijie Yang, Hongyin Luo, Jiawei Zhao, Tri Dao†, and Ravi Netravali†
    In Proceedings of the Conference on Language Modeling, 2026
    Code Website
  3. arXiv
    Geometry Guided Self-Consistency for Physical AI
    Yinwei Dai, Zhuofu Chen, Lijie Yang, and Ravi Netravali
    2026
    Code Website
  4. Tech Report
    CaveAgent: Transforming LLMs into Stateful Runtime Operators
    Maohao Ran, Zhenglin Wan, Cooper Lin, Yanting Zhang, Hongyu Xin, Hongwei Fan, Yibo Xu, Beier Luo, Yaxin Zhou, and 14 more authors
    2026
    Code Website
  5. MLSys
    Event Tensor: A Unified Abstraction for Compiling Dynamic Megakernel
    Hongyi Jin, Bohan Hou, Guanjie Wang, Ruihang Lai, Jinqi Chen, Zihao Ye, Yaxing Cai, Yixin Dong, Xinhao Cheng, and 12 more authors
    In Proceedings of the Conference on Machine Learning and Systems, 2026
    Website
  6. ICML
    Less Is More: Training-Free Sparse Attention with Global Locality for Efficient Reasoning
    Lijie Yang*, Zhihao Zhang*, Arti Jain, Shijie Cao, Baihong Yuan, Yiwei Chen, Zhihao Jia, and Ravi Netravali
    In Proceedings of International Conference on Machine Learning, 2026
    Code Website

2025

  1. Tech Report
    Beyond Context Limits: Subconscious Threads for Long-Horizon Reasoning
    Hongyin Luo, Nathaniel Morgan, Tina Li, Derek Zhao, Ai Vy Ngo, Philip Schroeder, Lijie Yang, Assaf Ben-Kish, Jack O’Brien, and 1 more author
    2025
    Website
  2. IoT
    From Machine Learning-Based to LLM-Enhanced: An Application-Focused Analysis of How Social IoT Benefits from LLMs
    Lijie Yang, and Runbo Su
    IEEE Internet of Things Journal, Apr 2025
    Website
  3. ICLR
    TidalDecode: Fast and Accurate LLM Decoding with Position Persistent Sparse Attention
    Lijie Yang*, Zhihao Zhang*, Zhuofu Chen, Zikun Li, and Zhihao Jia
    In proceedings of International Conference on Learning Representations, Apr 2025
    Code Website

2024

  1. LCN
    Blocking-Waived Estimation: Improving the Worst-Case End-To-End Delay Analysis in Switched Ethernet
    Lijie Yang, Théo Docquier, Ludovic Thomas, and Ye-Qiong Song
    In Proceedings of Local Computer Networks, Apr 2024
    Code Website
  2. ICML
    Accelerating Retrieval-augmented Language Model Serving with Speculation
    Zhihao Zhang, Alan Zhu, Lijie Yang, Yihua Xu, Lanting Li, Phitchaya Mangpo Phothilimthana, and Zhihao Jia
    In Proceedings of International Conference on Machine Learning, Apr 2024
    Code Website
  3. ASPLOS
    SpecInfer: Accelerating Large Language Model Serving with Tree-based Speculative Inference and Verification
    Xupeng Miao*, Gabriele Oliaro*, Zhihao Zhang*, Xinhao Cheng*, Zeyu Wang, Zhengxin Zhang, Rae Ying Yee Wong, Alan Zhu, Lijie Yang, and 6 more authors
    In Proceedings of the 29th ACM International Conference on Architectural Support for Programming Languages and Operating Systems, Volume 3, Apr 2024
    Code Website

2023

  1. HAL
    Technical Report: Worst-case Delay Analysis: a Simulation-based Comparison between Flow Aggregation and CPA
    Lijie Yang
    Jan 2023
    PDF