Machine learningLLMs & Text
SliceGPT: Compress Large Language Models by Deleting Rows and Columns
Compressing Language Models: The paper introduces SliceGPT, a post-training sparsification scheme for large language models that reduces the network's embedding dimension, maintains high performance, reduces inference computation, and reveals computational invariance in transformer networks.
Featured in No. 35 on 30 Jan 2024 · 4 days after release · 466 citations today · published in International Conference on Learning Representations
- Released
- 26 Jan 2024
- First featured
- No. 35 · 30 Jan 2024
- Citations (Semantic Scholar)
- 466
- Influential citations
- 66
- Published in
- International Conference on Learning Representations
- Shares when featured
- 20
- Identifier
- arXiv:2401.15024
Citations and venue from Semantic Scholar (ODC-BY), refreshed weekly. Summary: Quant Letter (CC BY 4.0).