The paper proposes memory recurrent units (MRUs) and a bistable implementation (BMRU) that combine persistent memory from nonlinear RNNs with the parallelizable training of state-space models.
Learning Finite State Machines With Self-Clustering Recurrent Networks,
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
-
Parallelizable memory recurrent units
The paper proposes memory recurrent units (MRUs) and a bistable implementation (BMRU) that combine persistent memory from nonlinear RNNs with the parallelizable training of state-space models.