📄 Paper of the Day2026-07-18
Layerskip: Enabling Early Exit Inference And Self-speculative Decoding2024 · 234 cites · Model Architecture, Training Techniques, Efficiency
We present LayerSkip, an end-to-end solution to speed-up inference of large language models (LLMs). First, during training we apply layer dropout, with low dropout rates for earlier layers and higher dropout rates for la…