2 papers
cs.AR2026
A Reconfigurable and Representation-Adaptive ISA-Based Architecture for Efficient DNN Acceleration
Vasilis Sakellariou, Vassilis Paliouras, Ioannis Kouretas +2
Domain-specific hardware accelerators provide significantly higher performance and energy efficiency for deep neural network (DNN) workloads than general-purpose processors, but of…
cs.LG2024
Hybrid Dynamic Pruning: A Pathway to Efficient Transformer Inference
Ghadeer Jaradat, Mohammed Tolba, Ghada Alsuhli +4
In the world of deep learning, Transformer models have become very significant, leading to improvements in many areas from understanding language to recognizing images, covering a…