1 paper
Tony Zhang, Rickard Brännvall
This work explores optimizing transformer-based language models by integrating model compression techniques with inhibitor attention, a novel alternative attention mechanism. Inhib…