conftrace_

monarch matrice

4 papers

Explore in graph

Co-occurring keywords · papers mentioning both

sequence length scaling (1) block tensor-train (btt) (1) compute-efficient models (1) sparse parameterization (1) sum blocks (1) golubeva, neyshabur and gur-ari (1) kronecker-product weights (1) sparseness (1) unstructured sparse-weight mlps (1) initialization scale (1)

Papers

Scaling Probabilistic Circuits via Monarch Matrices ICML 2025 Compute Better Spent: Replacing Dense Layers with Structured Matrices ICML 2024 Understanding MLP-Mixer as a wide and sparse MLP ICML 2024 Monarch Mixer: A Simple Sub-Quadratic GEMM-Based Architecture NeurIPS 2023