Block Pruning is a model compression technique that belongs to the family of structured pruning methods, distinguished by its removal of entire blocks of parameters—such as weight matrices or layers—based on predefined criteria to achieve computational efficiency without retraining.
28
Publications
798
Citations
114
Authors
43
Institutions
Publications per year
2017–2026
21
Parents
Publications
28
Structured Pruning of Large Language Models
2020 · 88 citations · Full text
Faster BlockMax WAND with Variable-sized Blocks
Antonio Mallia, Giuseppe Ottaviano, Elia Porciani et al. · 2017 · 64 citations
1–5 of 28
Authors
114
- WN
6
h-index
- YW
6
h-index
- BR
6
h-index
- XL
4
h-index
- ZL
4
h-index
1–5 of 114
Institutions
43
Universidad del Noreste
Tampico, Mexico
3 h-index
15 publications
809 citations
William & Mary
Williamsburg, United States
3 h-index
6 publications
302 citations
Xiamen University
Xiamen, China
2 h-index
6 publications
277 citations
Institut National des Sciences Appliquées de Strasbourg
Strasbourg, France
1 h-index
1 publications
22 citations
Institut de Recherche en Informatique et Systèmes Aléatoires
Rennes, France
1 h-index
1 publications
22 citations
Peking University
Beijing, China
1 h-index
1 publications
50 citations
Universidade de Brasília
Brasília, Brazil
1 h-index
3 publications
84 citations
Birkbeck, University of London
London, United Kingdom
1 h-index
1 publications
58 citations
1–8 of 43
Venues
- arXiv (Cornell University)
Repository · 6 publications · 100 citations
6
h-index
- Proceedings of the AAAI Conference on Artificial Intelligence
Conference · 3 publications · 112 citations
3
h-index
- ACM Transactions on Parallel Computing
TOPC · 1 publications · 28 citations
1
h-index
- Proceedings of the ACM on Programming Languages
PACMPL · 1 publications · 18 citations
1
h-index
- IEEE Transactions on Network and Service Management
TNSM · 1 publications · 25 citations
1
h-index
1–5