77 citations · 126 across the 12 of their papers we have counts for
13 papers
Programming AMD XDNA NPUs with Open-source Compiler Tools: A FlashAttention Case Study
Erwei Wang, Ephrem Wu, Victor J. B. Jung +4
Spatial NPUs such as AMD XDNA place compute tiles beside small local memories and leave data movement between them to software. Mapping a multi-stage workload onto such a device is…
From Human Guidance to Autonomy: Agent Skill System for End-to-End LLM Deployment on Spatial NPUs
Jiajie Li, Erwei Wang, Zhiru Zhang +1
Spatial neural processing units (NPUs) provide an energy-efficient platform for edge LLM inference, but efficiently deploying an LLM end-to-end on such hardware remains labor-inten…
Practical Formal Verification for MLIR Programs
Emily Tucker, Louis-Noël Pouchet, Erika Hunhoff +2
Optimizing compilers have become a cornerstone for high-performance program generation in research and industry. Optimizations, including those implemented manually by a user and t…
KernelCraft: Benchmarking for Agentic Close-to-Metal Kernel Generation on Emerging Hardware
Jiayi Nie, Haoran Wu, Yao Lai +9
New AI accelerators with novel instruction set architectures (ISAs) often require developers to manually craft low-level kernels, a time-consuming and error-prone process that does…
From Loop Nests to Silicon: Mapping AI Workloads onto AMD NPUs with MLIR-AIR
Erwei Wang, Samuel Bayliss, Andra Bisca +19
General-purpose compilers abstract away parallelism, locality, and synchronization, limiting their effectiveness on modern spatial architectures. As modern computing architectures…
Logic Shrinkage: Learned FPGA Netlist Sparsity for Efficient Neural Network Inference
Erwei Wang, James J. Davis, Georgios-Ilias Stavrou +3
FPGA-specific DNN architectures using the native LUTs as independently trainable inference operators have been shown to achieve favorable area-accuracy and energy-accuracy tradeoff…