Finetuned Language Models Are Zero-Shot Learners ​

← Back to topic

Authors: Hyung Won Chung, Le Hou, Shayne Longpre, Barret Zoph, Yifei Tang, William Fedus, Eric Li, et al.
Year: 2021
Journal: arXiv
DOI: 10.48550/arXiv.2109.01652
Publisher: https://arxiv.org/abs/2109.01652

Keywords: instruction tuning, zero-shot, flan

Abstract ​

We show that instruction tuning dramatically improves zero-shot performance.

Cite this paper ​

@misc{flant52021,
  title  = {Finetuned Language Models Are Zero-Shot Learners},
  author = {Hyung Won Chung, Le Hou, Shayne Longpre, Barret Zoph, Yifei Tang, William Fedus, Eric Li, et al.},
  year   = {2021},
  journal = {arXiv},
  doi    = {10.48550/arXiv.2109.01652},
  url    = {https://doi.org/10.48550/arXiv.2109.01652},
}

Source files ​

• metadata.json
• paper.bib