The neural code for syntax is claimed to be a hybrid of symbolic hierarchical structure (the ROSE architecture) and statistical linear prediction, linked by phase-amplitude coupling.
In summary, current connectionist models of syntax typically exhibit a strong bias for syntactically simple linear-distance units (Lakretz et al
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CL 1years
2024 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Shadow of the (Hierarchical) Tree: Reconciling Symbolic and Predictive Components of the Neural Code for Syntax
The neural code for syntax is claimed to be a hybrid of symbolic hierarchical structure (the ROSE architecture) and statistical linear prediction, linked by phase-amplitude coupling.