On the Convergence of Encoder-only Shallow Transformers.

Yongtao Wu Fanghui Liu Grigorios G. Chrysos Volkan Cevher

Published in: CoRR (2023)

Keyphrases

bit rate
information extraction
question answering
low complexity
rate distortion
natural language processing
convergence rate
data sets
information systems
knowledge base
multiscale
macroblock
faster convergence
syntactic parsing
deep knowledge