dot-resize-optimal-transport-compression

【效率与协作】Compress LLMs by 20-30% in width while preserving functionality through optimal transport-based neuron merging. Instead of discarding neurons, redistribute their signal to retained neurons via learned transport maps. Use when you need to reduce model size with minimal accuracy loss and measurable computational speedup.

独立 GitHub 快照 · 自动分类 · 以原作者文档为准