Yujun Lin | NVIDIA

About Me

I am currently a research scientist at NVIDIA. I finished my PhD at MIT HAN Lab of MIT EECS, advised by Prof. Song Han. I received my B.Eng. in Electronic Engineering from Tsinghua University.

My research area is efficient deep learning, with a special focus on the co-design of algorithm, system and hardware for foundation models (diffusion models, LLMs, etc).

News

[05/2025] “SANA 1.5: Efficient Scaling of Training-Time and Inference-Time Compute in Linear Diffusion Transformer” is accepted to ICML’25.
[05/2025] “Sparse Video-Gen: Accelerating Video Diffusion Transformers with Spatial-Temporal Sparsity” is accepted to ICML’25.
[02/2025] “QServe: W4A8KV4 Quantization and System Co-design for Efficient LLM Serving” is accepted to MLSys’25.
[02/2025] “LServe: Efficient Long-Sequence LLM Serving with Unified Sparse Attention” is accepted to MLSys’25.
[01/2025] “SVDQuant: Absorbing Outliers by Low-Rank Components for 4-Bit Diffusion Models” is accepted to ICLR’25 (Spotlight).
[01/2025] “SANA: Efficient High-Resolution Image Synthesis with Linear Diffusion Transformers” is accepted to ICLR’25 (Oral).
[10/2024] “LEGO: Spatial Accelerator Generation and Optimization for Tensor Applications” is accepted to HPCA’25.
[05/2023] “VideoTime³: A 40-uJ/frame 38 FPS Video Understanding Accelerator With Real-Time DiffFrame Temporal Redundancy Reduction and Temporal Modeling” is accepted to IEEE Solid-State Circuits Letters.
[08/2022] “TorchSparse: Efficient Point Cloud Inference Engine” is accepted to MLSys’22.
[04/2022] “QuantumNAS: Noise-Adaptive Search for Robust Quantum Circuits” is accepted to HPCA’22.
[10/2021] Seleted as DAC Young Fellow 2021.
[07/2021] “PointAcc: Efficient Point Cloud Accelerator” is accepted to MICRO’21.
[06/2021] Got into the 2021 Qualcomm Innovation Fellowship winners list
[02/2021] “NAAS: Neural Accelerator Architecture Search” is accepted to DAC’21.

Selected Publications

ICML

Sparse Video-Gen: Accelerating Video Diffusion Transformers with Spatial-Temporal Sparsity

Haocheng Xi, Shuo Yang, Yilong Zhao, Chenfeng Xu, Muyang Li, Xiuyu Li, Yujun Lin, Han Cai, Jintao Zhang, Dacheng Li, Jianfei Chen, Ion Stoica, Kurt Keutzer, Song Han

ICML, 2025.

Paper Code

ICML

SANA 1.5: Efficient Scaling of Training-Time and Inference-Time Compute in Linear Diffusion Transformer

Enze Xie, Junsong Chen, Yuyang Zhao, Jincheng Yu, Ligeng Zhu, Chengyue Wu, Yujun Lin, Zhekai Zhang, Muyang Li, Junyu Chen, Han Cai, Bingchen Liu, Daquan Zhou, Song Han

ICML, 2025.

Paper Code Project Page

ICLR

SVDQuant: Absorbing Outliers by Low-Rank Components for 4-Bit Diffusion Models

Muyang Li*, Yujun Lin*, Zhekai Zhang*, Tianle Cai, Xiuyu Li, Junxian Guo, Enze Xie, Chenlin Meng, Jun-Yan Zhu, Song Han.

ICLR, 2025.

Paper Code Project Page Spotlight Presentation

HPCA

LEGO: Spatial Accelerator Generation and Optimization for Tensor Applications

Yujun Lin*, Zhekai Zhang*, Song Han.

HPCA, 2025.

Project Page

ICLR

SANA: Efficient High-Resolution Image Synthesis with Linear Diffusion Transformers

Enze Xie, Junsong Chen, Junyu Chen, Han Cai, Yujun Lin, Zhekai Zhang, Muyang Li, Yao Lu, Song Han

ICLR, 2025.

Paper Code Project Page Oral Presentation

MLSys

QServe: W4A8KV4 Quantization and System Co-design for Efficient LLM Serving

Yujun Lin*, Haotian Tang*, Shang Yang*, Zhekai Zhang, Guangxuan Xiao, Chuang Gan, Song Han.

MLSys, 2025.

Paper Code Project Page

MLSys

LServe: Efficient Long-Sequence LLM Serving with Unified Sparse Attention

Shang Yang*, Junxian Guo*, Haotian Tang, Qinghao Hu, Guangxuan Xiao, Jiaming Tang, Yujun Lin*, Zhijian Liu, Yao Lu, Song Han

MLSys, 2025.

Project Page

SSCL

VideoTime³: A 40-uJ/frame 38 FPS Video Understanding Accelerator With Real-Time DiffFrame Temporal Redundancy Reduction and Temporal Modeling

Miaorong Wang, Yujun Lin, Zhekai Zhang, Ji Lin, Song Han, Anantha P Chandrakasan.

IEEE Solid-State Circuits Letters, 2023.

Paper Project Page

MLSys

TorchSparse: Efficient Point Cloud Inference Engine

Haotian Tang*, Zhijian Liu*, Xiuyu Li*, Yujun Lin, Song Han.

MLSys, 2022.

Paper Code Project Page

HPCA

QuantumNAS: Noise-Adaptive Search for Robust Quantum Circuits

Hanrui Wang, Yongshan Ding, Jiaqi Gu, Zirui Li, Yujun Lin, David Z. Pan, Frederic T. Chong, Song Han.

HPCA, 2022.

Paper Code Project Page Oral Presentation

MICRO

PointAcc: Efficient Point Cloud Accelerator

Yujun Lin, Zhekai Zhang, Haotian Tang, Hanrui Wang, Song Han

MICRO, 2021.

Paper Project Page Oral Presentation

DAC

NAAS: Neural Accelerator Architecture Search

Yujun Lin*, Mengtian Yang*, Song Han

DAC, 2021.

Paper Project Page Oral Presentation

ECCV

Searching Efficient 3D Architectures with Sparse Point-Voxel Convolution

Haotian Tang*, Zhijian Liu*, Shengyu Zhao, Yujun Lin, Ji Lin, Hanrui Wang, Song Han

ECCV, 2020.

Paper Code Project Page

NeurIPS

MCUNet: Tiny Deep Learning on IoT Devices

Ji Lin, Wei-Ming Chen, Yujun Lin, John Cohn, Chuang Gan, Song Han

NeurIPS, 2020.

Paper Code Project Page Spotlight Presentation

ICLR

Lite Transformer with Long-Short Range Attention

Zhanghao Wu*, Zhijian Liu*, Ji Lin, Yujun Lin, Song Han

ICLR, 2020.

Paper Code Project Page

NeurIPS

Point-Voxel CNN for Efficient 3D Deep Learning

Zhijian Liu*, Haotian Tang*, Yujun Lin, Song Han

NeurIPS, 2019.

Paper Code Project Page Spotlight Presentation

CVPR

HAQ: Hardware-Aware Automated Quantization with Mixed Precision

Kuan Wang*, Zhijian Liu*, Yujun Lin*, Ji Lin, Song Han

CVPR, 2019.

Paper Code Project Page Oral Presentation

DAC

Long live TIME: improving lifetime for training-in-memory engines by structured gradient sparsification

Yi Cai*, Yujun Lin*, Lixue Xia, Xiaoming Chen, Song Han, Yu Wang, Huazhong Yang

DAC, 2018.

Paper

ICLR

Deep Gradient Compression: Reducing the communication bandwidth for distributed training

Yujun Lin, Song Han, Huizi Mao, Yu Wang, William J. Dally

ICLR, 2018.

Paper Code Project Page

Services

Conference Reviewer: ICLR, ICML, NeurIPS
Journal Reviewer: TPAMI, TACD, TACO, TAI, TCSVT, TETC, TECS, TKDE, TNNLS, VISI, etc.