BingoCGN, a scalable and efficient graph neural network accelerator that enables inference of real-time, large-scale graphs through graph partitioning, has been developed by researchers at the ...

BingoCGN employs cross-partition message quantization to summarize inter-partition message flow, which eliminates the need for irregular off-chip memory access and utilizes a fine-grained structured ...

Graphs are widely used to represent complex relationships in everyday applications such as social networks, bioinformatics, and recommendation ...