3 papers
cs.NI2025
Analyzing Communication Predictability in LLM Training
Wenxue Li, Xiangzhou Liu, Yuxuan Li +9
Effective communication is essential in distributed training, with predictability being one of its most significant characteristics. However, existing studies primarily focus on ex…
physics.app-ph2025
LightCode: Compiling LLM Inference for Photonic-Electronic Systems
Ryan Tomich, Zhizhen Zhong, Dirk Englund
The growing demand for low-latency, energy-efficient inference in large language models (LLMs) has catalyzed interest in heterogeneous architectures. While GPUs remain dominant, th…
cs.NI2025
MixNet: A Runtime Reconfigurable Optical-Electrical Fabric for Distributed Mixture-of-Experts Training
Xudong Liao, Yijun Sun, Han Tian +13
Mixture-of-Expert (MoE) models outperform conventional models by selectively activating different subnets, named experts, on a per-token basis. This gated computation generates dyn…