Divisive normalization in a recurrent network yields low-rank slow manifolds that support continuous working memory, whereas subtractive inhibition shatters them under time-varying input.
Title resolution pending
1 Pith paper cite this work, alongside 166 external citations. Polarity classification is still indexing.
1
Pith paper citing it
166
external citations · OpenAlex
fields
q-bio.NC 1years
2026 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Divisive Normalization Shapes Low-Rank Slow Manifolds for Continuous Working Memory
Divisive normalization in a recurrent network yields low-rank slow manifolds that support continuous working memory, whereas subtractive inhibition shatters them under time-varying input.