3 papers
cs.AI2026
TorchTPU - A Hybrid Eager/Compiled Architecture for Native PyTorch Execution on TPU
Claudio Basile, Michael Voznesensky, Gunhyun Park +51
TorchTPU is an open-source collaboration between Google and Meta that introduces a native and principled PyTorch backend for Google TPUs. It enables seamless out-of-the-box model e…
cs.AI2026
MaxKernel: Agentic Kernel Generation for TPUs
Shangkun Wang, Nina Cai, Charles Hoong +7
Designing and authoring high-performance custom kernels for accelerators is a complex task that requires deep hardware-level expertise. Large Language Models (LLM) can be leveraged…
cs.AI2026
Agentic Framework for Deep Learning workload migration via In-Context Learning
Qiyue Liang, Steven Ingram, George Vanica +4
Translating deep learning models from PyTorch's flexible, object-oriented design to JAX's functional, stateless setup is usually a manual and error-prone task. Automated migration…