XTC: ExtremeCompressionforPre-trained TransformersMadeSimpleandEfficient

Neural Information Processing Systems 

Asaresult,wefindoutthatprevious baselines for ultra-low bit precision quantization are significantly under-trained. Based on our study,we propose asimple yet effectivecompression pipeline for extreme compression, named XTC.

Similar Docs  Excel Report  more

TitleSimilaritySource
None found