๐ŸŽ“How I Study AIHISA
๐Ÿ“–Read
๐Ÿ“„Papers๐Ÿ“ฐBlogs๐ŸŽฌCourses
๐Ÿ’กLearn
๐Ÿ›ค๏ธPaths๐Ÿ“šTopics๐Ÿ’กConcepts๐ŸŽดShorts
๐ŸŽฏPractice
โฑ๏ธCoach๐ŸงฉProblems๐Ÿง Thinking๐ŸŽฏPrompts๐Ÿง Review
SearchSettings
How I Study AI - Learn AI Papers & Lectures the Easy Way

Papers1

AllBeginnerIntermediateAdvanced
All SourcesarXiv
#orthogonal Procrustes

COMPOT: Calibration-Optimized Matrix Procrustes Orthogonalization for Transformers Compression

Intermediate
Denis Makhov, Dmitriy Shopkhoev et al.Feb 16arXiv

COMPOT is a training-free way to shrink Transformer models while keeping their smarts.

#Transformer compression#orthogonal dictionary learning#orthogonal Procrustes