Pith. sign in

Efficient scaling of large language models with mixture of experts and 3d analog in-memory computing

1 Pith paper cite this work. Polarity classification is still indexing.

1 Pith paper citing it

fields

cs.LG 1

years

2025 1

verdicts

REJECT 1

representative citing papers

Composing Linear Layers from Irreducibles

cs.LG · 2025-07-15 · reject · novelty 6.0

A rotor-based layer built from bivector exponentials approximates LLM attention projections with O(log^2 d) parameters and competitive downstream performance.

citing papers explorer

Showing 1 of 1 citing paper.

  • Composing Linear Layers from Irreducibles cs.LG · 2025-07-15 · reject · none · ref 6

    A rotor-based layer built from bivector exponentials approximates LLM attention projections with O(log^2 d) parameters and competitive downstream performance.