Kinman Lei
Qihao Zhang
Mingliang Tang
Mingshu Zhai
Kinman Lei
Jidong Zhai
RoMeo: Mitigating Dual-dimensional Outliers with Rotated Mixed Precision Quantization.
273-287
2026
PPoPP
https://doi.org/10.1145/3774934.3786419
conf/ppopp/2026
db/conf/ppopp/ppopp2026.html#ZhangTZLZ26
Runxin Zhong
Yuyang Jin 0001
Chen Zhang 0001
Kinman Lei
Shuangyu Li
Jidong Zhai
FlashTensor: Optimizing Tensor Programs by Leveraging Fine-grained Tensor Property.
183-196
2025
PPoPP
https://doi.org/10.1145/3710848.3710864
conf/ppopp/2025
db/conf/ppopp/ppopp2025.html#ZhongJ0LLZ25
Haoyu Yang
Zan Zong
Yuyang Jin 0001
Kinman Lei
Jiaao He
Qigang Yang
Jidong Zhai
UltraAttn: Efficiently Parallelizing Attention through Hierarchical Context-Tiling.
1769-1784
2025
SC
https://doi.org/10.1145/3712285.3759894
conf/sc/2025
db/conf/sc/sc2025.html#YangZ0LHYZ25
Kezhao Huang
Siqi Zhu
Mingshu Zhai
Liyan Zheng 0001
Kinman Lei
Jiaao He
Yuyang Jin 0001
Jidong Zhai
mTuner: Accelerating Parameter-Efficient Fine-Tuning on Multi-GPU Servers with Elastic Tensor.
105-121
2025
USENIX ATC
https://www.usenix.org/conference/atc25/presentation/huang-kezhao
conf/usenix/2025
db/conf/usenix/usenix2025.html#HuangZZ0LH0Z25
Kinman Lei
Yuyang Jin 0001
Mingshu Zhai
Kezhao Huang
Haoxing Ye
Jidong Zhai
PUZZLE: Efficiently Aligning Large Language Models through Light-Weight Context Switch.
127-140
2024
USENIX ATC
https://www.usenix.org/conference/atc24/presentation/lei
conf/usenix/2024
db/conf/usenix/usenix2024.html#LeiJZHYZ24
Jiaao He
Kezhao Huang
Yuyang Jin 0001
Shuangyu Li
Mingliang Tang
Haoyu Yang
Qigang Yang
Haoxing Ye
Jidong Zhai
Mingshu Zhai
Chen Zhang 0001
Qihao Zhang
Liyan Zheng 0001
Runxin Zhong
Siqi Zhu
Zan Zong