Kinman Lei

According to our database, Kinman Lei authored at least 5 papers between 2024 and 2026.

Collaborative distances:

• Dijkstra number of four.
• Erdős number of four.

2026

RoMeo: Mitigating Dual-dimensional Outliers with Rotated Mixed Precision Quantization.

Qihao Zhang, Mingliang Tang, Mingshu Zhai, Kinman Lei, Jidong Zhai

Proceedings of the 31st ACM SIGPLAN Annual Symposium on Principles and Practice of Parallel Programming, 2026

2025

mTuner: Accelerating Parameter-Efficient Fine-Tuning on Multi-GPU Servers with Elastic Tensor.

Kezhao Huang, Siqi Zhu, Mingshu Zhai, Liyan Zheng, Kinman Lei, Jiaao He, Yuyang Jin, Jidong Zhai

Proceedings of the 2025 USENIX Annual Technical Conference, 2025

UltraAttn: Efficiently Parallelizing Attention through Hierarchical Context-Tiling.

Haoyu Yang, Zan Zong, Yuyang Jin, Kinman Lei, Jiaao He, Qigang Yang, Jidong Zhai

Proceedings of the International Conference for High Performance Computing, 2025

FlashTensor: Optimizing Tensor Programs by Leveraging Fine-grained Tensor Property.

Runxin Zhong, Yuyang Jin, Chen Zhang, Kinman Lei, Shuangyu Li, Jidong Zhai

Proceedings of the 30th ACM SIGPLAN Annual Symposium on Principles and Practice of Parallel Programming, 2025

2024

PUZZLE: Efficiently Aligning Large Language Models through Light-Weight Context Switch.

Kinman Lei, Yuyang Jin, Mingshu Zhai, Kezhao Huang, Haoxing Ye, Jidong Zhai

Proceedings of the 2024 USENIX Annual Technical Conference, 2024