Abstract: We propose reparameterized refocusing convolution (RefConv) as a replacement for regular convolutional layers, which is a plug-and-play module to improve the performance without any ...

Abstract: Graph Neural Networks (GNNs) have been gaining more attention due to their excellent performance in modeling various graph-structured data. However, most of the current GNNs only consider ...

A discrete-time signal is a sequence of numbers defined only at integer values of time. Addition combines two signals sample by sample. y[n] = x₁[n] + x₂[n] Used in mixing and superposition of signals ...

In this paper, we tackle the high computational overhead of transformers for lightweight image super-resolution. (SR). Motivated by the observations of self-attention's inter-layer repetition, we ...