DYNAMAX adds early-exit classifiers to Mamba and Transformer LLMs and uses Mamba blocks as classifiers, showing computational savings with limited performance loss.
Attention is All you Need,
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CL 1years
2025 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
DYNAMAX: Dynamic computing for Transformers and Mamba based architectures
DYNAMAX adds early-exit classifiers to Mamba and Transformer LLMs and uses Mamba blocks as classifiers, showing computational savings with limited performance loss.