conftrace_

feed-forward skipping

1 paper

Explore in graph

Co-occurring keywords · papers mentioning both

efficient generation (1) knowledge-intensive generation (1) autoregressive decoding (1) inference acceleration (1) layer skipping (1) early exit (1) model compression (1)

Papers

FFN-SkipLLM: A Hidden Gem for Autoregressive Decoding with Adaptive Feed Forward Skipping EMNLP 2024