Characterizing the Accuracy-Efficiency Trade-off of Low-rank Decomposition in Language Models

Chakshu Moar, Faraz Tahmasebi, Michael Pellauer, Hyoukjun Kwon. Characterizing the Accuracy-Efficiency Trade-off of Low-rank Decomposition in Language Models. In IEEE International Symposium on Workload Characterization, IISWC 2024, Vancouver, BC, Canada, September 15-17, 2024. pages 194-209, IEEE, 2024. [doi]

Authors

Chakshu Moar

This author has not been identified. Look up 'Chakshu Moar' in Google

Faraz Tahmasebi

This author has not been identified. Look up 'Faraz Tahmasebi' in Google

Michael Pellauer

This author has not been identified. Look up 'Michael Pellauer' in Google

Hyoukjun Kwon

This author has not been identified. Look up 'Hyoukjun Kwon' in Google