Session Chair: Lisa Wu Wills (Duke Univ.)

Mosaic: Exploiting Instruction-Level Parallelism on Deep Learning Accelerators with iTex Tessellation
Jianxing Xu (University of Science and Technology of China,SKL of Processors, Institute of Computing Technology, Chinese Academy of Sciences), Yuanbo Wen (SKL of Processors, Institute of Computing Technology, Chinese Academy of Sciences), Zikang Liu (SKL of Processors, Institute of Computing Technology, Chinese Academy of Sciences,University of Chinese Academy of Sciences), Ruibai Xu (University of Science and Technology of China,SKL of Processors, Institute of Computing Technology, Chinese Academy of Sciences), Tingfeng Ruan (SKL of Processors, Institute of Computing Technology, Chinese Academy of Sciences,University of Chinese Academy of Sciences), Jun Bi (SKL of Processors, Institute of Computing Technology, Chinese Academy of Sciences), Rui Zhang (SKL of Processors, Institute of Computing Technology, Chinese Academy of Sciences), Di Huang (SKL of Processors, Institute of Computing Technology, Chinese Academy of Sciences), Xinkai Song (SKL of Processors, Institute of Computing Technology, Chinese Academy of Sciences), Yifan Hao (SKL of Processors, Institute of Computing Technology, Chinese Academy of Sciences), Xing Hu (SKL of Processors, Institute of Computing Technology, Chinese Academy of Sciences), Zidong Du (SKL of Processors, Institute of Computing Technology, Chinese Academy of Sciences), Chongqing Zhao (Tencent), Jie Jiang (Tencent), Qi Guo (SKL of Processors, Institute of Computing Technology, Chinese Academy of Sciences)
Paper

DynaX: Sparse Attention Acceleration with Dynamic X:M Fine-Grained Structured Pruning
Xiao Xiong (College of Computer Science, Chongqing University), Zhaorui Chen (College of Computer Science, Chongqing University), Yue Liang (College of Computer Science, Chongqing University), Minghao Tian (College of Computer Science, Chongqing University), Jiaxing Shang (College of Computer Science, Chongqing University), Jiang Zhong (College of Computer Science, Chongqing University), Dajiang Liu (College of Computer Science, Chongqing University)
Paper

Accelerating Retrieval-Augmented Generation
Derrick Quinn (Cornell University), Mohammad Nouri (Cornell University), Neel Patel (Cornell University), John Salihu (University of Kansas), Alireza Salemi (UMass Amherst), Sukhan Lee (Samsung Electronics), Hamed Zamani (UMass Amherst), Mohammad Alian (Cornell University)
Paper

GUST: Graph Edge-Coloring Utilization for Accelerating Sparse Matrix Vector Multiplication
Armin Gerami (Computer Science, University of Maryland), Bahar Asgari (Computer Science, University of Maryland)
Paper

RASSM: Residue-based Acceleration of Single Sparse Matrix Computation via Adaptive Tiling
Anirudh Jain (Georgia Institute of Technology), Pulkit Gupta (Georgia Institute of Technology), Thomas M. Conte (Georgia Institute of Technology)
Paper