Abstract: In the field of AI inference chips, quantization can effectively reduce the parameter size and computational cost. Additionally, the design of hardware acceleration modules can significantly ...
Some results have been hidden because they may be inaccessible to you
Show inaccessible results
Feedback