About Me

I am a Ph.D. candidate in Computer Science and Engineering at Pennsylvania State University, advised by Prof. Mahmut Kandemir and Prof. Kamesh Madduri. I previously recieved a B.S. in Biomechatronics and an M.S. in Electrical Engineering from National Taiwan University.

I am passionate about Machine Learning Systems, with a focus on areas such as AutoML, deep learning compilers, and large-scale distributed training. My work primarily involves LLMs, GNNs, and DLRMs. Currently, I am working on improving parallelization strategies for LLM and Graph Transformer training, which are critical for scaling the next generation of multi-modal LLMs.

Publications

Enhancing Graph Transformer Training through Adaptive Graph Parallelism
Jun-Liang Lin, Kamesh Madduri, Mahmut Taylan Kandemir
IEEE International Parallel and Distributed Processing Symposium Workshops (IPDPSW), 2025  

Parallelization Strategies for DLRM Embedding Bag Operator on AMD CPUs
K. Nair et al., including Jun-Liang Lin
IEEE Micro, 2024  
[Paper]

Thorough Characterization and Analysis of Large Transformer Model Training At-Scale
Scott Cheng, Jun-Liang Lin, Murali Emani, Siddhisanket Raskar, Sam Foreman, Zhen Xie, Venkatram Vishwanath, Mahmut Taylan Kandemir
Proceedings of the ACM on Measurement and Analysis of Computing Systems (SIGMETRICS), 2024  
[Paper]

Quantization for Bayesian Deep Learning: Low-Precision Characterization and Robustness
Jun-Liang Lin, Ranganath Krishnan, Keyur Ruganathbhai Ranipa, Mahesh Subedar, Vrushabh Sanghavi, Meena Arunachalam, Omesh Tickoo, Ravishankar Iyer, Mahmut Taylan Kandemir
IEEE International Symposium on Workload Characterization (IISWC), 2023.  
[Paper] [Code]

ezGeno: an automatic model selection package for genomic data analysis
Jun-Liang Lin*, Tsung-Ting Hsieh*, Yi-An Tung*, Xuan-Jun Chen, Yu-Chun Hsiao, Chia-Lin Yang, Tyng-Luh Liu, Chien-Yu Chen
Bioinformatics, 2022  
[Paper] [Code]

The Maximum a Posteriori Estimation of DARTS
Jun-Liang Lin*, Yi-Lin Sung*, Cheng-Yao Hong*, Han-Hung Lee, and Tyng-Luh Liu
IEEE International Conference on Image Processing (ICIP), 2021  
[Paper]

Communication-Efficient Separable Neural Network for Distributed Inference on Edge Devices
Jun-Liang Lin, and Sheng-De Wang
arXiv preprint arXiv:2111.02489, 2021  
[Paper]

Experiences

Intern | Google
May 2025 - August 2025

Intern | Qualcomm
May 2024 - August 2024

Intern | Intel
May 2022 - April 2023

Research Assistant | Academia Sinica
September 2019 - August 2021

Awards

Student Travel Grant, SIGMETRICS 2024
Student Travel Grant, IISWC 2023
Graduate Research Fellowship, NTU, 2018
Professor Tomotake Takasaka Scholarship, NTU, 2016
Rong‑Zunn Wang Culture and Education Foundation Scholarship, NTU, 2015
Presidential Award, NTU, 2013, 2014, 2015

Services

Reviewer
International Conference on Learning Representations (ICLR)
Annual Conference on Neural Information Processing Systems (NeurIPS)
International Conference on Machine Learning (ICML)
IEEE Transactions on Computers
ACM Transactions on Intelligent Systems and Technology