2 papers
cs.CV2026
FTerViT: Fully Ternary Vision Transformer
Szymon Ruciński, Pietro Bonazzi, Engin Türetken +3
Ternary Vision Transformers offer substantial model compression, however state-of-the-art methods only ternarize the encoder layers, leaving patch embeddings, LayerNorm parameters,…
cs.NE2022
Optimizing the Consumption of Spiking Neural Networks with Activity Regularization
Simon Narduzzi, Siavash A. Bigdeli, Shih-Chii Liu +1
Reducing energy consumption is a critical point for neural network models running on edge devices. In this regard, reducing the number of multiply-accumulate (MAC) operations of De…