Active Instruction Tuning: Improving Cross-Task Generalization by Training on Prompt Sensitive Tasks
Top Cited Papers
Open Access
- 1 January 2023
- conference paper
- Published by Association for Computational Linguistics (ACL)
- p. 1813-1829
Abstract
Po-Nien Kung, Fan Yin, Di Wu, Kai-Wei Chang, Nanyun Peng. Proceedings of the 2023 Conference on Empirical Methods in Natural Language Processing. 2023.Keywords
All Related Versions
This publication has 18 references indexed in Scilit:
- Self-Instruct: Aligning Language Models with Self-Generated Instructions2023
- Unnatural Instructions: Tuning Language Models with (Almost) No Human Labor2023
- What Makes Instruction Learning Hard? An Investigation and a New Challenge in a Synthetic Environment2022
- ZeroPrompt: Scaling Prompt-Based Pretraining to 1,000 Tasks Improves Zero-Shot Generalization2022
- Fine-tuned Language Models are Continual Learners2022
- InstructDial: Improving Zero and Few-shot Generalization in Dialogue through Instruction Tuning2022
- What to Pre-Train on? Efficient Intermediate Task Selection2021
- Datasets: A Community Library for Natural Language Processing2021
- Evaluating the Values of Sources in Transfer Learning2021
- Dataset Cartography: Mapping and Diagnosing Datasets with Training Dynamics2020