works on
activation sparsity 1feed-forward networks 1large language models 1model compression 1training-free methods 1

From the 1 of 1 linked paper with an AI index.

collaborators
Showing cs.LGShow all

1 paper · 1 filter

cs.LG2026

Prox: Training-Free FFN Activation Sparsity via Approximate Intermediate-Channel Salience in LLMs

Jinyi Liu, Wei Chen, Pengyu Chen +4

The paper introduces Prox, a training-free framework that sparsifies feed‑forward network activations in large language models by approximating intermediate‑channel salience, enabl…