92 A Parallel Scan Algorithm in the Tensor Core Unit Model  
Anastasios Zouzias and William McColl

178 Improved Algorithms for Monotone Moldable Job Scheduling using Compression and Convolution    
Kilian Grage, Klaus Jansen and Felix Ohnesorge 

   
56 TrainBF: High-Performance DNN Training Engine using BFloat16 on AI Accelerators    
Zhen Xie, Siddhisanket Raskar, Murali Emani and Venkatram Vishwanath