CSAT applies compressed sensing ideas to Transformer attention by projecting keys and values into low dimensions and decoding outputs with learned sparse recovery, claiming linear complexity with benchmark performance near full attention.
Title resolution pending
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CV 1years
2025 1verdicts
REJECT 1representative citing papers
citing papers explorer
-
CS-VLM: Compressed Sensing Attention for Efficient Vision-Language Representation Learning
CSAT applies compressed sensing ideas to Transformer attention by projecting keys and values into low dimensions and decoding outputs with learned sparse recovery, claiming linear complexity with benchmark performance near full attention.