A monotone head-gating mechanism conditions transformer attention on a budget, enabling one checkpoint to trade attention cost for accuracy and produce measured CPU speedups.
Adaptive attention span in transformers
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.LG 1years
2026 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
Budgeted Attention Allocation: Cost-Conditioned Compute Control for Efficient Transformers
A monotone head-gating mechanism conditions transformer attention on a budget, enabling one checkpoint to trade attention cost for accuracy and produce measured CPU speedups.