2 papers
cs.DC2026
Moebius: Serving Mixture-of-Expert Models with Seamless Runtime Parallelism Switch
Shaoyu Wang, Yizhuo Liang, Jaeyong Song +2
Mixture-of-Experts (MoE) architectures scale large language models (LLMs) to hundreds of billions of parameters. Serving a single MoE model requires multiple GPUs operating in para…
cs.LG2026
Reversal Q-Learning
Aditya Oberai, Seohong Park, Sergey Levine
Iterative generative modeling techniques, such as flow matching, provide powerful tools to model complex behaviors for effective offline reinforcement learning (RL). In this work,…