conftrace_

router l2 norm

1 paper

Explore in graph

Co-occurring keywords · papers mentioning both

computational requirements (1) pruning experts (1) sparse mixture-of-experts (1) fine-tuned model (1) model size (1)

Papers

A Provably Effective Method for Pruning Experts in Fine-tuned Sparse Mixture-of-Experts ICML 2024