Session Chair: Baris Kasikci (UW)

Einsum Trees: An Abstraction for Optimizing the Execution of Tensor Expressions
Alexander Breuer (Friedrich Schiller University Jena), Mark Blacher (Friedrich Schiller University Jena), Max Engel (Friedrich Schiller University Jena), Joachim Giesen (Friedrich Schiller University Jena), Alexander Heinecke (Intel Corporation), Julien Klaus (Friedrich Schiller University Jena), Stefan Remke (Friedrich Schiller University Jena)
Paper

Optimizing Deep Learning Inference Efficiency through Block Dependency Analysis
Zhanyuan Di (SKLP, Institute of Computing Technology, CAS,University of Chinese Academy of Sciences), Leping Wang (SKLP, Institute of Computing Technology, CAS), En Shao (SKLP, Institute of Computing Technology, CAS,University of Chinese Academy of Sciences), Zhaojia Ma (SKLP, Institute of Computing Technology, CAS,University of Chinese Academy of Sciences), Ziyi Ren (SKLP, Institute of Computing Technology, CAS,University of Chinese Academy of Sciences), Feng Hua (SKLP, Institute of Computing Technology, CAS,University of Chinese Academy of Sciences), Lixian Ma (SKLP, Institute of Computing Technology, CAS,University of Chinese Academy of Sciences), Jie Zhao (Hunan University), Guangming Tan (SKLP, Institute of Computing Technology, CAS,University of Chinese Academy of Sciences), Ninghui Sun (SKLP, Institute of Computing Technology, CAS,University of Chinese Academy of Sciences)
Paper

Pruner: A Draft-then-Verify Exploration Mechanism to Accelerate Tensor Program Tuning
Liang Qiao (University of Science and Technology of China), Jun Shi (University of Science and Technology of China), Xiaoyu Hao (University of Science and Technology of China), Xi Fang (University of Science and Technology of China), Sen Zhang (University of Science and Technology of China), Minfan Zhao (University of Science and Technology of China), Ziqi Zhu (University of Science and Technology of China), Junshi Chen (University of Science and Technology of China), Hong An (University of Science and Technology of China), Xulong Tang (University of Pittsburgh), Bing Li (NIO), Honghui Yuan (NIO), Xinyang Wang (NIO)
Paper

Relax: Composable Abstractions for End-to-End Dynamic Machine Learning
Ruihang Lai (Carnegie Mellon University), Junru Shao (OpenAI), Siyuan Feng (Shanghai Jiao Tong University), Steven Lyubomirsky (NVIDIA), Bohan Hou (Carnegie Mellon University), Wuwei Lin (OpenAI), Zihao Ye (University of Washington), Hongyi Jin (Carnegie Mellon University), Yuchen Jin (Hyperbolic), Jiawei Liu (University of Illinois Urbana-Champaign), Lesheng Jin (Hyperbolic), Yaxing Cai (NVIDIA), Ziheng Jiang (ByteDance), Yong Wu (NVIDIA), Sunghyun Park (NVIDIA), Prakalp Srivastava (Netflix), Jared Roesch (NVIDIA), Todd C. Mowry (Carnegie Mellon University), Tianqi Chen (Carnegie Mellon University,NVIDIA)
Paper

Towards End-to-End Optimization of LLM-based Applications with Ayo
Xin Tan (The Chinese University of Hong Kong), Yimin Jiang (Unaffiliated), Yitao Yang (The Chinese University of Hong Kong), Hong Xu (The Chinese University of Hong Kong)
Paper