GRASP: Replace Redundant Layers with Adaptive Singular Parameters for Efficient Model Compression
Open Access
- 1 January 2025
- conference paper
- Published by Association for Computational Linguistics (ACL)
- p. 26344-26359
Abstract
Kainan Liu, Yong Zhang, Ning Cheng, Zhitao Li, Shaojun Wang, Jing Xiao. Proceedings of the 2025 Conference on Empirical Methods in Natural Language Processing. 2025.Keywords
All Related Versions
This publication has 8 references indexed in Scilit:
- ShortGPT: Layers in Large Language Models are More Redundant Than You Expect2025
- LaCo: Large Language Model Pruning via Layer Collapse2024
- The defeat of the Winograd Schema ChallengeArtificial Intelligence, 2023
- Dynamic Low-rank Estimation for Transformer-based Language Models2023
- CommonsenseQA: A Question Answering Challenge Targeting Commonsense Knowledge2019
- BoolQ: Exploring the Surprising Difficulty of Natural Yes/No Questions2019
- ChID: A Large-scale Chinese IDiom Dataset for Cloze Test2019
- RACE: Large-scale ReAding Comprehension Dataset From Examinations2017