Reproduce “A 25-$μ$s/inf Event-driven Graph Neural Network Processor with Spatiotemporal Caching and Spline Convolution for Ultra-low-latency AI at the Edge” on OpenResearch
Turn your favorite coding agents into research agents that review literature, develop hypotheses, and run experiments.
Universal app for Apple silicon and Intel. Requires macOS 11 or later.
About this paper
Adrian Kneip, Martin Lefebvre, Daniel Gehrig, Victoria Catalán Pastor, Davide Scaramuzza, Marian Verhelst, Charlotte Frenkel
ETHEREAL accelerates event-driven graph neural network processing for dynamic vision sensors by processing graph neighbors in parallel and using separate memory strategies for time-aware and spatial-only layers. On DSEC’s 640×480 input, it measures 25.6 μs and 1.7 μJ per event inference at 0.95 V with mixed 4/8-bit precision. This is a chip-level workload result, not measured real-world end-to-end detection accuracy.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.