Active Instruction Tuning: Improving Cross-Task Generalization by Training on Prompt Sensitive Tasks

Po-Nien Kung, Fan Yin, Di Wu, Kai-Wei Chang, Nanyun Peng


Abstract
Instruction tuning (IT) achieves impressive zero-shot generalization results by training large language models (LLMs) on a massive amount of diverse tasks with instructions. However, how to select new tasks to improve the performance and generalizability of IT models remains an open question. Training on all existing tasks is impractical due to prohibiting computation requirements, and randomly selecting tasks can lead to suboptimal performance. In this work, we propose active instruction tuning based on prompt uncertainty, a novel framework to identify informative tasks, and then actively tune the models on the selected tasks. We represent the informativeness of new tasks with the disagreement of the current model outputs over perturbed prompts. Our experiments on NIV2 and Self-Instruct datasets demonstrate that our method consistently outperforms other baseline strategies for task selection, achieving better out-of-distribution generalization with fewer training tasks. Additionally, we introduce a task map that categorizes and diagnoses tasks based on prompt uncertainty and prediction probability. We discover that training on ambiguous (prompt-uncertain) tasks improves generalization while training on difficult (prompt-certain and low-probability) tasks offers no benefit, underscoring the importance of task selection for instruction tuning.
Anthology ID:
2023.emnlp-main.112
Volume:
Proceedings of the 2023 Conference on Empirical Methods in Natural Language Processing
Month:
December
Year:
2023
Address:
Singapore
Editors:
Houda Bouamor, Juan Pino, Kalika Bali
Venue:
EMNLP
SIG:
Publisher:
Association for Computational Linguistics
Note:
Pages:
1813–1829
Language:
URL:
https://aclanthology.org/2023.emnlp-main.112
DOI:
10.18653/v1/2023.emnlp-main.112
Bibkey:
Cite (ACL):
Po-Nien Kung, Fan Yin, Di Wu, Kai-Wei Chang, and Nanyun Peng. 2023. Active Instruction Tuning: Improving Cross-Task Generalization by Training on Prompt Sensitive Tasks. In Proceedings of the 2023 Conference on Empirical Methods in Natural Language Processing, pages 1813–1829, Singapore. Association for Computational Linguistics.
Cite (Informal):
Active Instruction Tuning: Improving Cross-Task Generalization by Training on Prompt Sensitive Tasks (Kung et al., EMNLP 2023)
Copy Citation:
PDF:
https://aclanthology.org/2023.emnlp-main.112.pdf
Video:
 https://aclanthology.org/2023.emnlp-main.112.mp4