About

I am a senior Ph.D. student in Electrical and Computer Engineering at Rice University, advised by Prof. Tong (Tony) Geng. I previously spent two years with the same group at the University of Rochester, and received my B.S. and M.S. from Zhejiang University.

My research sits at the intersection of computer architecture and machine learning systems. I build hardware–software co-designed systems that let physical dynamical systems perform the computation digital accelerators struggle with — spanning graph learning, differential equation solving, and large language model training and inference.

I am always happy to talk about nature-powered computing, accelerator design, and efficient learning systems. Feel free to reach out by email.

  • Dynamical-System / Nature-Powered Computing
  • Efficient Training and Inference of Large Language Models
  • Computer Architecture & Hardware–Software Co-Design
  • Scientific Computing Acceleration

Publications

* denotes equal contribution.

2026

  • MICRO

    BITE: Boosting LLM Training via Sandwich-Bite Dataflow and Dynamic Subspace Auto-Alignment

    R. Song, T. Wang, G. Sun, C. Wu, Q. Wang, D. Liu, S. Kondguli, Y. Hao, A. Li, T. Geng

    59th IEEE/ACM International Symposium on Microarchitecture (MICRO)

  • ISCA

    DS-ISA: Instruction Set Architecture for Dynamical System Units

    C. Wu, R. Song, C. Liu, T. Geng, A. Li

    53rd ACM/IEEE International Symposium on Computer Architecture (ISCA)

  • ICLR

    Zeros can be Informative: Masked Binary U-Net for Image Segmentation on Tensor Cores

    C. Wu, R. Song, S. Kondguli, T. Geng, A. Li

    International Conference on Learning Representations (ICLR)

  • ICML

    Solving Time-Dependent Differential Equations with Physical Dynamical Systems

    C. Liu, Y. Chen, R. Song, W. Huang, C. Wu, D. Kong, Y. N. Wu, K. Yang, A. Li, T. Geng

    International Conference on Machine Learning (ICML)

2025

  • ISCA

    DS-TPU: Dynamical System for on-Device Lifelong Graph Learning with Nonlinear Node Interaction

    R. Song*, C. Wu*, C. Liu, P. Haghi, A. Li, M. Huang, T. Geng

    52nd ACM/IEEE International Symposium on Computer Architecture (ISCA)

  • ICLR

    DS-LLM: Leveraging Dynamical Systems to Enhance Both Training and Inference of Large Language Models

    R. Song, C. Liu, C. Wu, A. Li, D. Liu, Y. Wu, T. Geng

    Thirteenth International Conference on Learning Representations (ICLR)

  • MICRO

    DS-TIDE: Harnessing Dynamical Systems for Efficient Time-Independent Differential Equation Solving

    C. Liu, C. Wu, R. Song, G. Sun, Y. Wu, Y. Chen, A. Li, T. Geng

    58th IEEE/ACM International Symposium on Microarchitecture (MICRO)

  • ICML

    An Expressive and Self-Adaptive Dynamical System for Efficient Function Learning

    C. Liu, C. Wu, R. Song, A. Li, Y. Wu, T. Geng

    42nd International Conference on Machine Learning (ICML)

  • ICLR

    InstaTrain: Adaptive Training via Ultra-Fast Natural Annealing within Dynamical Systems

    C. Liu, R. Song, C. Wu, P. Haghi, T. Geng

    Thirteenth International Conference on Learning Representations (ICLR)

  • DAC

    DM-Tune: Quantizing Diffusion Models with Mixture-of-Gaussian Guided Noise Tuning

    P. Haghi, A. Falahati, Z. Azad, C. Wu, R. Song, C. Liu, A. Li, T. Geng

    62nd ACM/IEEE Design Automation Conference (DAC)

  • ASP-DAC

    Nature-GL: A Revolutionary Learning Paradigm Unleashing Nature's Power in Real-World Spatial-Temporal Graph Learning

    C. Liu, C. Wu, R. Song, Y. Chen, A. Li, M. Huang, T. Geng

    30th Asia and South Pacific Design Automation Conference (ASP-DAC)

2024

  • ISCA

    DS-GL: Advancing Graph Learning via Harnessing Nature's Power within Scalable Dynamical Systems

    R. Song*, C. Wu*, C. Liu, A. Li, M. Huang, T. Geng

    51st ACM/IEEE International Symposium on Computer Architecture (ISCA)

  • ICLR

    NP-GL: Extending Power of Nature from Binary to Real-valued Graph Learning in Real World

    C. Wu, R. Song, C. Liu, Y. Yang, A. Li, M. Huang, T. Geng

    Twelfth International Conference on Learning Representations (ICLR)

  • LoG

    NP-NDS: A Nature-Powered Nonlinear Dynamical System for Power Grid Forecasting

    C. Wu, R. Song, C. Liu, Y. Wang, Y. Chen, A. Li, D. Liu, Y. Wu, M. Huang, T. Geng

    Third Learning on Graphs Conference (LoG)

  • TCHES

    A Highly-efficient Lattice-based Post-Quantum Cryptography Processor for IoT Applications

    Z. Ye, R. Song, H. Zhang, D. Chen, R. C.-C. Cheung, K. Huang

    IACR Transactions on Cryptographic Hardware and Embedded Systems (TCHES)

2022

  • TCSVT

    A Reconfigurable Convolution-in-Pixel CMOS Image Sensor Architecture

    R. Song, K. Huang, Z. Wang, H. Shen

    IEEE Transactions on Circuits and Systems for Video Technology (TCSVT), 32(10)

2021

  • PIERS

    A Near-array Convolution Computing Scheme Based on WSe2 Photodiode

    L. Cai, K. Huang, R. Song, H. Shen

    Photonics & Electromagnetics Research Symposium (PIERS)

Awards

  • 2025 Graduate Teaching AwardUniversity of Rochester

Education

  • Jan 2026 – Present Ph.D. in Electrical and Computer Engineering Rice University, Houston, TX
  • Sep 2023 – Dec 2025 Ph.D. in Electrical and Computer Engineering University of Rochester, Rochester, NY Transferred to Rice with Prof. Geng's group
  • Sep 2020 – Jun 2023 M.S. in Electronic Science and Technology Zhejiang University, Hangzhou, China Overall rank: 1/14
  • Sep 2016 – Jul 2020 B.S. in Electronic and Information Engineering Zhejiang University, Hangzhou, China