Scaling & shifting your features: A new baseline for efficient model tuning

D Lian, D Zhou, J Feng, X Wang - Advances in Neural …, 2022 - proceedings.neurips.cc
Existing fine-tuning methods either tune all parameters of the pre-trained model (full fine-
tuning), which is not efficient, or only tune the last linear layer (linear probing), which suffers …
被引用次数：152 相关文章

Scaling & Shifting Your Features: A New Baseline for Efficient Model Tuning

D Lian, D Zhou, J Feng, X Wang - arXiv preprint arXiv:2210.08823, 2022 - arxiv.org
Existing fine-tuning methods either tune all parameters of the pre-trained model (full fine-
tuning), which is not efficient, or only tune the last linear layer (linear probing), which suffers …

Scaling & Shifting Your Features: A New Baseline for Efficient Model Tuning

D Lian, D Zhou, J Feng, X Wang - arXiv e-prints, 2022 - ui.adsabs.harvard.edu
Existing fine-tuning methods either tune all parameters of the pre-trained model (full fine-
tuning), which is not efficient, or only tune the last linear layer (linear probing), which suffers …

Scaling & shifting your features: a new baseline for efficient model tuning

D Lian, D Zhou, J Feng, X Wang - Proceedings of the 36th International …, 2022 - dl.acm.org
Existing fine-tuning methods either tune all parameters of the pre-trained model (full fine-
tuning), which is not Efficient, or only tune the last linear layer (linear probing), which suffers …

Scaling & Shifting Your Features: A New Baseline for Efficient Model Tuning

D Lian, Z Daquan, J Feng, X Wang - Advances in Neural …, 2022 - openreview.net
Existing fine-tuning methods either tune all parameters of the pre-trained model (full fine-
tuning), which is not efficient, or only tune the last linear layer (linear probing), which suffers …
[PDF] neurips.cc

[PDF][PDF] Scaling & Shifting Your Features: A New Baseline for Efficient Model Tuning

D Lian, D Zhou, J Feng, X Wang - papers.neurips.cc
Existing fine-tuning methods either tune all parameters of the pre-trained model (full fine-
tuning), which is not efficient, or only tune the last linear layer (linear probing), which suffers …