We are excited to announce that two papers from PADSYS Lab have been accepted and presented at SC 2025:

  • "HPC-R1: Characterizing R1-like Large Reasoning Models on HPC"
  • "DPAR: High-Performance, Secure, and Scalable Differential Privacy-based AllReduce"

These works explore emerging directions in large-scale AI systems and secure distributed training.

Congratulations to all authors!

Go PADSYS!