Oct 2025AI Multi-Accelerator PyTorch Serving With NxD Inference and vLLMYahav Biran & Liangfu Chen, Amazon
Oct 2025AI mxfp8, mxfp4, nvfp4 formats and applications in PyTorchVasily Kuznetsov & Driss Guessous, Meta
Oct 2025AI Optimizing Long-Tail and MoE Challenges in Reinforcement Learning with SGLangChenyang Zhao, UCLA
Oct 2025AI PyTorch APIs for High Performance MoE Training and InferenceD. Vega-Myhre; Ke Wen & N. Gimelshein
Oct 2025AI PyTorch Symmetric Memory: A New Programming Paradigm for Distributed AIKe Wen & Chien-Chin Huang
Oct 2025AI Rethinking the Transformer: Toward Native Multimodal ArchitecturesBowen Peng, Nous Research
Oct 2025AI Scaling Inference of O(10K)-length Sequence Recommendation Models Using...S. Joshi & K. Rajesh