Functorial Neural Architectures from Higher Inductive Types 事件

Functorial Neural Architectures from Higher Inductive Types arXiv:2603.16123v2 Announce Type: replace-cross Abstract: Neural networks often learn the parts of a task but fail on novel combinations of those parts. We argue that this failure is architectural: a decoder generalizes compositionally only when it respects the algebraic laws of the task, i.e. when it descends from freely generated sequences to the quotient determined by those laws. We make this principle constructive by compiling High