SOURCE-LINKED INTELLIGENCE

GeoPair: Geometry-Preserving Cross-Layer Factorization for Training-Free Transformer Compression

arXiv · AI, language, vision and robotics · article · Sep 22, 2026 · UTC

Transformer architectures exhibit cross-layer redundancies, yet post-training compression pipelines typically optimize layers in isolation or rely on heuristic grouping strategies that disregard layer-specific activation geometries. We introduce a principled, training-free framework that sequentially optimizes cross-layer weight pairings and shared-dictionary factorizations. Rather than forcing weights of adjacent layers to share a basis or heuristically merging activation statistics, our approach identifies structurally compatible projections and learns a shared representation that better pre

Read original source ↗ Open in workspace

recordType

paper

region

Global