conftrace_

2:4 sparsity

4 papers

Explore in graph

Co-occurring keywords · papers mentioning both

model compression (3) sparse training (2) structured sparsity (2) weight recovery (1) large language model (1) dense fine-tuning (1) spatial similarity (1) information restoration (1) flip rate (1) masked decay term (1)

Papers

Accelerating Transformer Pre-training with 2:4 Sparsity ICML 2024 S-STE: Continuous Pruning Function for Efficient 2:4 Sparse Pre-training NeurIPS 2024 WRP: Weight Recover Prune for Structured Sparsity ACL 2024 Towards Fully Sparse Training: Information Restoration with Spatial Similarity AAAI 2022